1
1
THE
MACH
THAT CHANGED THE
BASED ON TI1
MRSSACHUSmS
INSTITUTE
0-
THE
VA
THAT
i.lli
CHANGED THE
BEBíBiNObOGY
5-MILLION-DOLL AR
S-YEAR
STUW
JAMES i?WOMACK
DANIEL T. JONES
DANIEL ROOS
ON TWE
FUTURE
OF BWE
RA W S O N A S S O C I A T E S
New York
COLLIER MACMILLAN CANADA
AUTOMQBlLE
TORONTO
MAXWELL MACMILLAN INTERNATIONAL
NEW Y O R K
OXFORD
SINGAPORE
SYDNEY
Cupyl-ight O 1990 by lames P. Wornack, Ilanicl T. Juncs, Daniel Kuos, and Doriria
Samtnons Cai-pentcr
Al1 rights rcscrved. No part of this bouk m- be reprudiiced or transmittc<I in any
forrn or by any rneaiis, ~ l e c t i o n i cor rnechanicai. including photucopying. recording
r>rby üny iníorrnatioii storage and rrtricval system, ivithout prrmission in writing
iiom thc Publislicr.
Rawson Associatcs
Macmilkin Publishing Cumpany
866 Third Arcnue, Nrw York, N.Y. IOOLZ
Cullier Macrnillan Canada, Inc.
Coiiier Macrnillan Canada.. Inc.
1200 Eglinton Avrnuc Easi
Suitc 200
Don Milis, Ontario M3C 3N1
Milimilla" boriks are available at special discounts far bulk purchasrs fur sales
prornotions, prcmiums, fund-raising, or educational use.
For detaiis, contact:
Packaged by Rapid Transcript, a divisiun of March Tenth, Inc.
Designed by Stanlry S . DratelFolio Graphics Co. Inc.
1 0 9 8 7 6 5 4 3 2 1
Printcd in thc Unitrd Statcs of Arnerica
1
THE INDUSTRY OF INDUSTRIES IN
TRANSITION
3
17
~~
ISBN: 0-892.56-350-8
LCC numbcr: 89-063284
Special Sales Director
Macmillan Publishing Conipany
866 Ttrird Avenue
New York, N.Y. 10022
BEFORE YOU BEGIN THlS BOOK
THE ORlGlNS OF LEAN PRODUClTlOM
17
2
THE RlSE AND FALL OF MASS PRODUCTION
27
3
THE RlSE OF LEAN PRODUCTION
48
THE ELEMENTS OF LEAN PRQOUGTION
4
S
RUNNING THE FACTORY
DESlGNlNG THE CAR
V
vii
ACKNOWLEDGMENTS
gram but speaking to a general audience-reflects a successful
melding ol two distinct cultures.
I t lvould al1 have been for naught except for the extraordinary
generosity of our IMVP research affiliates in sharing their knowledge freely. While three of us are listed as auihors because we put
ihe words on the page, this has truly been a group effort by
individuals from many backgrounds and countries. We have tried
to acknowledge tbeir contributions fully at appropriate points jn
the text and end notes. The reader should always bear in mind
that, lilre the development of a new car in a "lean" auto company,
this has been a truly collective endeavor.
DANIEL
Roos, Director, IMVP
T. JONES,European Director, IMVP
DANIEL
JAMES
P. WOMACK,
Research Director, IMVP
1
THE
MACH
1 THAT CHANGED THE
traveling the world surveying ninety auto assembly plants in
fiffien countries, in what we believe is the most comprehensive
industrial survey ever undertaken in any industry.
T\VOadditional MBA students at MIT, Antony Sherjff an
Kentaro Nobeoka, provided insight for our product-development
studies, through case studies of the product-development process
based on their previous work as product planners at Chrysler and
Mazda, respectively.
A mere listing of these naines shows an additional feature
our work that we felt was essential-to develop a completel
internationai team of researchers, with the language and cultur
skills to understand production methods in different countries
and an eagerness to explain iheir findings to colleagues [rom very
diffel-ent backgrounds. These researchers (who are listed in APpendix B) were not primarily stationed at MIT and were not
primarily American. Ratlier, we developed a worldwide team with
no geographic center and no one nationality in the majority.
To be taken seriously both inside and outside the motorvehiele industry we needed to be independent. Therefore, we
determined Lo raise the $5 million we needed through contribu.
tions from many car companies, components suppliers, and governmcnts. (The thirty-six organizations ultimately contributing
to the IMVP are listed in Appendix A,) We limited contributions
from individual companies and governments to 5 percent
the
$S-million total and placed al1 the Funds in a single account, so
that no sponsor could influence the direction of our work by
earmarking its contribution Sor a special purpose. we were also
careful Lo raise funds in equal amounts in North America, western
Europe, and Japan, so that we would not be subject to national or
regional pressures in our conclusions.
For our researchers to succeed, they would need extensive
access to motor-vehicle companies across the world, from the
factory floor to the executive suite. We therefore made it elear to
potential sponsors that their most valuable contribution would
not be money but rather the time given by their employees to
answer our questions. In every case these companies have been
even more open than we had hoped. We have been tmly amazed
by the spirit of professionalism in this industry, which has moved
managers in the worst facilities and weakest companies to share
their ~ r o b l e m sfrankly, and managers in tlie best plants and
strongest companies to explain their secrets candidly.
Finally, to succeed in our work we were determined to devise
feedback mechanisms where we could explain our findings
stry, governments, and unions and gain their reactions for
ual benefit. We did this in three ways.
t, we held an annual meeting Eor the liaison Person from
meeting we went over the previous year's
r, ~t
detail, asking for criticism and for suggestions about
s for our research.
d, we held an annual policy fomm at a diflerent locand the world-~iagara-on-the-Lake in Canada, Como in
capulco in Mexic-to
present our findings to senior execd government officials from the sponsoring c o m ~ a n i e s
overnments,plus interested observers from labor unioi's
he financial community. These private meetings provided an
tunity for senior leaders ol this industry to discuss the real
moving the world from mass to lean production,
publicity and the need for public posturing.
the
attending the IMVP policy forums are listed in A P P ~ ~ inally, we've conducted several hundred private briefings for
anies, governments, and unions. For example, our factory
ice team eonducted a seminar at each of the ninety assembly
we "isited as part of the IMVP World AssembIy PlanL
ey, I~ these seminars, we reviewed worldwide performance.
sed the performance of the plant we were visiting, and
ained the reasons that plant might lag in world-class ~ e r f o r m . we also conducted briefings h r corporate management
ds, union executive committees, government ministries. an*
rs in the investment community, in each case explaining the
rences between mass production and lean with ideas on how
onvert to lean production.
have now spent five years exploring the differences between
ass production and leati production in one enormous i n d u s t ~ .
have been both insiders with access to vast amounts of propriry information and daily contact with i n d u s t r ~leaders, and
,,"tsiders with a broad perspectivc, often very critical, o* existing
practices. ~n this process we've become convinced that the prin-
12
13
THE MACHlNE THAT CHANGED THE WORU)
Manufacturers around the world are now trying to embrace
lean production, but they're finding the going rough. The cornpa.
nies that first mastered this system were al1 headquartered in one
country-Japan. As lean production has spread to North Americ
and Western Europe under their aegis, trade wars and growing
resistance to foreign investment have fojlowed.
Today, we hear constantly that the world faces a massive
overcdpacity crisis--estimated by some industry executives at
more than 8 million units in excess of current world sales of about
50 miliion units.' This is. in fact, a misnorner. ~h~ Miorld has an
acute shortage of competitive lean-production capacity and a vast
glut of uncompetitive mass-production capacity. The crisis is
caused by the former threatening the Iatter.
Many Western companies now understand lean production,
~
and at least one is weil along the path to introducing it. H
SuPerimposing lean-production methods on existing mas,+prOduction systems causes great pain and dislocation. In the absenfe
of a crisis threatening the very survival of the company, only
limited progress seems to be possible.
General Motors is the most siriking example. This gigantic
company is still the world's largest industrial concern and was
without doubt the best at mass production, a system it helped lo
create. Now, in the age of lean production, it finds itsel[ with toa
manY managers. toa many workers, and too many plants. yet GM
has not yet faced a life-or-death crisis, as the Ford Motor company
did in the early 1980s, and thus it has not been able to change.)
This book is an effort tu ease the necessary transition from
mass production to lean. By focusing on the global auto industry,
we explain in simple, concreie terms what lean production is,
where it carne frorn, how it really works, and how it can spread to
al1 corners of the globe for everyone's mutual benefit,
Bu1 why should we care if world manufacturers jettison
decades of mass production to ernbrace lean production? Because
the adoption of lean production, as it inevitably spreads beyond
the auto industry, will change everything in almost every indus.
try-choices for consumers, the nature o1 work, the fortune of
companies, and, ultimately, the fate of nations,
What is lean production? Perhaps the best way to describe
this innovative production system is to coiltrast it with craft
production and mass pruduction, the two other methods humans
have devised to make things.
The craft producer uses highly skilled workers and simple but
~
exactly what tlie consumer asks for-one
to
ime, customfiimiture, works of decorative art, and a
love
ports cars provide current-day examples. we
craft production, but the problem with it is obvious:
method-as automobiles once were
uced by the
toa much for most ol us to afford. So mass
eveloped at the beginning of the iwenticth ceI1tive.
iass-producer uses narrowly skilled profcssiailals tu
ducts made by unskilled or semiskilled worlcers teilding
ingle-purpose machines. These churn out standardin very high volume. Because the machiner- costs
d is so intolerant of disruption, thc mass-pi-oducer
any buffers-xtra
supplies, extra workers, and exl.ra
to
assure
smooth
production.
~
~
~
~
, Because changing over a
even
more,
rhe
mass-producer
keeps standard
oduct
production for as long as possible. The result: The
er gets lower costs but at the expense of variety and by
of work methods that niost employees hnd boring a1ld
e lean producer, by contrast, combines the advantages uf
nd rnass production, while avoiding the high cost of the
and the rigidity of the latter. Toward this end, lean produmploy teams of multislcilled workers at al1 levels of thc
ization and use highly flexible, increasingly autornated maes to produce volumes of producís ir1 enorrnous v a r i e t ~ .
production (a term coined by IMVP researcher John
ik) is "leanu because it uses less of everything cumpared
mass pmduetion-half thc human effort in the factol-Y,hall
ufacturing space, half the investment in tools, half the
ing hours to develop a new product in half the time. Also,
quires keeping far less than half the needed inventory on sile,
lts in many fewer defects, and produces a grcater and ever
ing variety of products.
perhaps the most striking difference between mass prvducand lean production lies in their ultimate objcctives. m s s ,-,
enough,
ducers set a limited goal for themselves-"good
h translates into an acceptable number of defects, a maxiacceptable leve1 of inventaries, a narrow range of standardproducts. yo do better, they argue, xvouid Cost ioo much or
ed inherent human capabilities.
L~~~ producers, on ¡he other hand, set their sights expliciily
ever will, but the endless quest for perfection continues t
ate surprising twists.
For one, lean production changes how people work
sing ~ e a nproduction," we examine how
reading across the world and to other indus, lean production isn't spreading everY-
sowe'll
look at the barriers that are
es and countries from becoming lean. *nd
e ways leanness can be achieved.
,
anxiety about making costly mistakes.
Similarly, lean production changes the meaning of
sional careers. In the West, we are accustomed to think of
as a continua1 progression to ever higher levels of technica
work, the less you may know about a specific, narrow speci
that you can take with you to another company or to start a
business. What's more, many employees may find the lack
ual challenges, workers may feel they have reached a dead e
an early point in their career. The result: They hold back
book we provide the answers.
LEAN PRODUCT
o new idea springs hll-biown fronl a void.
om a set of conditions in xvbich olcl
rk. This was certainly true of lean
e country a t a specilic time because
al ideas for the industrial development of the country
e, to iinderstand lean production
igins fully, it is important to go much farther hack in
ct, back to the origins of the motor indusiry at the end
a t the craft origins of the industry in
and the transition to mass production around 1915,
countered problems it could not surtake pains to describe the mature system of mass
as it came to exist by the I920s, including iis strengths
nesses, because the system's weaknesses eventually bee source of inspiration for the next advance in industrial
ady to examine the genesis of lean
r it took root. We also summarize
developed lean production system as
the 1960s, a t a point long before the
2
In 1894, ihe Honorable Evelyn Wenry Ellis, a
Parliament, set
wealthv member OS the Enrlish
out to buy a car.' He didn't go to a car dealerthere lveren't any. Nor did he contact an English
automobile manulacturer-there weren't any of those either.
Instead, he visited the noted Paris machine-tool company of
Panhaid et Levassor and comrnissioned an automobile. Today,
P&L, as it was known, is remembered only by classic-cal- collectors and auto history buffs, but, in 1894, it was the world's leading
car company .2
It got its start-and a jump on other potential competitorswhen in 1887 Emile Levassor, the "L" of P&L, met Gottlieb
Daimler, the Iounder of rhe company that today builds the Mercedes-Benz. Levassor negotiaied a license to manufacture Daimler's new "high-speed" gasoline engine.
By the early 1890s, P&L was building severa1 hundred automobiles a year. The cars were designed according to the Systeme
Paizhard-meaning the engine was in the front, with passengers
seated in rows behind, and the motor drove the rear wheels.
When Ellis arrived at P&L, which was still primarily a manufacturer of metal-cutting saws rather than automobiles, he found
in place the elassic craft-produetion system. P&L's work force was
22
THE MACHlNE THATCHANGED JHE WORLD
ovenvhelmingly composed of skilled craftspeople who carefully
hand-built cars in small numbers.
These workers thoroughly understood mechanical design
principies and the materials with which they worked. What's
more, many were their own bosses, often serving as independent
contractors within the P&L plant or, more frequently, as independent rnachine-shop owners with whom the company contracted
for specific parts or components.
The two company founders, Panhard and Levassor, and their
immediate associates were responsible for talking to customers to
determine rhe vehicle's exact specifications, ordering the necessary parts, and assembling the final product. Much of the work,
though, including design and engineering, took place in individual craft shops scattered throughout Paris.
One oiour most basic assumptions in the age of mass production-ihat cost per unit falls dramatically as production volume
increases-was simply not true for craft-based P&L. If the company had tried to make 200,000 identical cars cach year, the cost
per car probably wouldn't have dipped much below the cost per
car of making ten.
What's more, P&L could never have made tivo-much less
200,000-identical cars, even if these were built to the same
blueprints. The reasons? P&L contractors didn't use a standard
gauging systeni, and the machine tools of the 1890s couldn't cut
hardened steel.
Instead, different contractors, using slightly different gauges,
made the parts. They then ran the parts through an oven to
harden their surfaces enough to withstand heavy use. However,
the parts frequently warped in the oven and needed further
macliiriing to regain their original shape.
When these parts eventually arrived at P&L's final assembly
hall, their specifications could best be described as approximate.
The Job of the skilled fitters in the hall \vas to take the first two
parts and file them down until they fit together perfectly.
Then they filed the third part until it fit the first two, and
so on until the whole vehicle-with its hundreds of parts-was
complete.
This sequential fitting produced what we today cal1 "dimensional creep." So, by the time the fitters reached the last part, the
total vehicle could differ significantly in dimensions from the car
on the next stand that was being built to the same biueprints.
Because P&L couldn't mass-produce identical cars, it didn't
JHE RlSE ANO FALL OFMASS PRODUCTION
23
try. Instead, it concentrated on tailoring each product to the
precise desires of individual buyers.
It also emphasized its cars' performance and their hand-fitted
which the gaps between individual parts were
mers Panhard was trying to woo, this pitch made
se wealthy customers employed chauffcurs and
ir personal staffs. Cost, driving ease, and simple
intenance weren't their primary concerns. Speed and customEvelyn Ellis was no doubt typical of P&L's clients. He didn't
t just any car; he wanted a car built to suit his precise needs
tastes. He was willing to accept P&L's basic chassis and
engine, he told the fir111's owners, but he wanted a special body
eonstructed by a Paris coachbuilder.
He also made a request to Levassor that would strike today's
auto manufacturer as preposterous: He asked that the transmission, brake, and engine controls be transferred frorn the right to
the left side of the car. (His reason wasn't that the English drove
on the left-in that case, moving the controls to the left side of the
vehicle was precisely the wrong thing to do. Besides, the steering
tiller remained in the middle. Rather, he presumably thought the
controls were more comfortable to use in that position.)
For P&L, Ellis's request probably seemed simple and reasonable. Since each part was made one at a time, it was a simple
matter to bend control rods to the left rather h a n the right and
to reverse the linkages. For today's mass-producer, this modification would require years-and millions or hundreds of millions of
dollars-to engineer. (American companies still offer no rightside-drive option on cars they se11 in drive-on-the-left Japan,
since they believe the cost of engineering the option would be
Once his automobile was finished, Ellis, accompanied by a
mechanic engaged for the purpose, tested it extensively on the
Paris sti-eets. For, unlike today's cars, the vehicle he had just
bought was in every sense a prototype. When he was satisfied that
his new car operated properly-quite likely after many trips back
to the P&L factory for adjustment-Ellis set off for England.
His arrival in June 1895 made history. Ellis became the first
person to drive an automobile in England. He traversed the fiftysix miles from Southampton to his country home in a mere 5
hours and 32 minutes-exclusive o€ stops-for an average speed
24
THE MACHINE THAT CHANGED THE WORLD
of 9.84 miles per hour. This speed was, in fact, flagrantly illegal,
since the limit in England for non-horse-drawn vehicles was a
sedate 4 miles per hour. But Ellis didn't intend to remain a
lawbreaker.
By 1896, he had taken the Parliamentary lead in repealing
the "flag law" that limited automotive speeds, and had organized
an Emancipation Run from London to Brighton, a trip on which
some cars even exceeded the new legal speed limit of 12 miles per
hour. Around this time, a number of English firms began to build
cars, signaling that the automotive age was spreading from its
origins in France to England in its march across the world.
Evelyn Ellis and P&L are worth remembering, despite the
subsequent failure of the Panhard firm and the crudeness of Ellis's
1894 auto (which found a home iii the Science Museum in London,
where you can see it today). Together, they perfectly summarize
the age of craft production in the motor industry.
In sum, craft production had the following characteristics:
A work force that was highly skilled iil design, machine
operations, and fitting. Most workers progressed through
an apprenticeship to a full set of craft skills. Many could
hope to run their own machine shops, becoming self-employed contractors to assembler firms.
Organizations that were extremely decentralized, although
concentrated within a single city. Most parts and much of
the vehicle's design came from small machine shops. The
system was coordinated by an ownerlentrepreneur in direct
contact with everyone involved-customers, employers, and
suppliers.
* The use of general-purpose machine tools to perform drilling, grinding, and other operations on metal and wood.
* A very low production volume-1,000 or fewer automobiles
ayear, only a few of which (fifty or fewer) were built to the
same design. And even among those fifty, no two were
exactly alike since craft techniques inherently produced
variations.
No company, of course, could exercise a monopoly over these
resources and characteristics, and Panhard et Levassor was soon
competing with scores of other companies, al1 producing vehicles
in a similar manner. By 1905, less than twenty years after P&L
produced the first commercially successful automobile, hundreds
THE RISEAND FALL OFMASS PRODUCTION
25
of companies in Wcstern Europe and North America were turning
out autos in small volumes using craft techniques.
The auto industry progressed to mass production aí'ter World
War 1, and P&L eventually f«undcred trying to ~ n a k ethe conversion. Yct, a number ol crafi-production firms have survived up to
the present. They continue to locus on tiny niches around the
upper, luxury end oS the market, populated with buyers wanting
a unique image and thc opportunity to deal directly with the
factory in ordering their vehicles.
Aston Rilartirl, for example, has produced fewer tlian 10,000
cars at its English workshop over the past sixty-fiue years and
currently turns out only one automobile each working day. Ir
survives by remaining small and exclusive, making a virtue of thc
high prices its craft-production tecliniques require. In its body
shop, for example, skilled panel beaters make the aluminum body
panels by pounding sheets oS aluminum with wooden mallets.
In the 1980s, as the pace of technological advances in the auto
industry has quicltened, Aston Martin and similar iirms have had
to ally themselves with the automotive giants (Ford, in Aston
Martin's case') in order to gain specialized expertise in areas
ranging from emission controls to crash safety. The cost of their
developing tliis expertise independently would have been simply
prohibitive.
In the 1990s, yet anotlier tlireat will emerge for these craft
firms as companies mastering lean production-led by the Japanese-begin to pursue their market niches, which were too small
and specializecl lor the mass-producers, such as Ford and GM,
ever to have successhlly attacked. For example, Honda has just
introduced its aluminum-bodied NS-X sports car, which is a
direct attack on Ferrari's niche in ultra-high-performance sports
cars. If these lean-production firms can cut design and manufacturing costs and improve on the produci quality offered by the
craft firms-and they probably can-the traditional craft producers will either have to adopt lean-production methods themselves or perish as a species after more than a century.
Nostalgists see Panhard and its competitors as the golden age
of auto production: Craftsmanship counted and companies gave
their full attention to individual consumers. Moreover, proud
craft worlters honed their skills and inany became independent
shop owners.
That's al1 true, but the drawbacks of craft production are
equally obvious in hindsight. Production costs were high and
THE MACHlNE THAT CHANGED THE WORLD
26
didn't drop with volume, which meant that only the rich could
afforci cars. In addition, because each car produced was, in effcct,
a prototype, consistency and reliability were elusive. (This, by the
way, is the same problem that plagues satellites and ibe U.S.
space shuttle, today's most prominent craft products.)
Car owners like Evelyn Ellis, or their chaufkurs and mechanics, had to provide their own on-the-road testing. In other words,
the system failed to provide product quality-in the lorm of
reliability and durabiiity rather than lots of leather os walnutbecause of the Iack of systematic testing.
Also fatal to the age, however, was the inability of thc small
independent shops, where most of the production work took place,
to develop new technologies. Individual craitsmen simply did not
have the resources to pursue fundamental innovations; real iechnological advance would have required systematic research rather
than just tinkering. Add these limitations together and it is clear,
in retrospect, that the industry was reaching a plateau when
Henry Ford came along. That is, as the general design of cars and
trucks began to converge on the now familiar four-~vheel,frontengine, internal-con~bustionvehicle we know today, the industq~
reached a premature maturity, fertile ground for a new production idea.
At this point, Henry Ford found a way to overcome the problems inherent in craft production. Ford's new techniques would
reduce costs dramatically while increasing product quality. Ford
called his innovative system n~assprodttction.4
MASS PRODUCTION
Ford's 1908 Model T was his twentieth design over a five-year
period that began with the production of the original Model A in
1903. With his Model T, Ford finally achieved two objectives. He
had a car that was designed for manufacture, as we would say
today, and that was, also in today's terms, user-friendly. Almost
anyone could drive and repair the car without a chauffeur or
mechanic. These two achievements laid the groundwork for the
revolutionary change in direction for the entire motor-vehicle
industry?
The key to mass production wasn't-as many people then and
THE RlSE AND FALL OF MASS PROUCTION
-
27
now belieuc-the riloving, or contiiiuous, asseinbly linc. R(i!/7ei., i!
ivus the col?zplete ~1i7d
C U I ~ . S ~ . $ ! ~ Iir~fe~li~liijieiibili!~
Z~
O ~ ~ L I ) . ~ ciizd
.S
IIZ~
sinzplicity ofafíaching iilcnz io i~cicilo ! / ~ eThese
i
were tlir maii~riacturing iiinovations that niade tlic assembly line possililc.
To achieve intcrcliangcability, Fbrd isisisicd thai ihe samc
gauging systern be uscd Sor cveiy part al1 tlie ivay ihruugh ihe
entire manuSacturing process. His iiisistencc on working-to-gauge
throughout \vas driven by his realization ol !he payofl' he ivo~ild
get in the Sorm olcavings on assenibiy costs. Remarltribiy, no one
eIse in the lledgiing inclustry had ligi~i-cd
out ilris criusc-and-eliec:;
so no orie else pur-sucd woi-lcirlg-tu-gnuge wiili Forcl's near-religious zeal.
Ford also henelittcd f'rom receiii advariccs in riiachiiie iools
able to work on puekurzici?ctimctals. The warpirig tliai occurred as
machincd parts \\rer-e hcing liardened had been tlic bane OS previous attempts to standardire parts. Oiice tlie warping problern wiis
solwd, Ford was able to develop innovative dcsigns thai rcduced
the number os parts ricecied and made tlicse parts easy io atiach.
For example, Ford's four-cylinder enzinc block consisted of a
single, complcx cactiiig. Competitors cast each cylincier separately and boltet! tlie Soui- together.
Taken rogethcr, intcrchangcahiliiy, simplicity, and case of
attachnient cave Ford ~rernendousiidvantages over- his cornpctition. For one, he could eliminate the skilled fitters who had always
formed the bullc of ercry assrmblcr's labor Force.
Ford's first efforts tu assemhle his cars, beginniiig in 1903,
involved setting up asseriibly stands on wliicli a whole car was
built, often by onc fitter. 111 1908, on ihe cve of tlic introduction oí'
the Model 1:a Ford assembler's average taslc cycle-the amoulit
of tiirre Iie \vol-ked be(orc repe~ctingihe samc operations-toialed
514 minutes, os 8.56 hours. Eacli worlter would asscinhle a largc
part of a car before nioving un to tlie nex:. For example, a worlter
might put al1 the niechanical paris-~vheels, springs, motor,
transmission, gcnerator-on the chassis, a set uf activities tliai
took a whole day to complete. The ;issenibler/litters pcriormed
the same set ol activities oler and over at their statioiiary asscmbly stands. They had to get the necessary parts, file tliem down so
they would fii (Ford hadn't yet acliievcd perfect interchangeability
of parts), thcn bol1 thesn ir>place.
The firsi step Ford toolr to rnake tliis process more efficieni
was to deliver :tic parts Lo each \\rorlc station. Now tlie asscmblers
could rcmain at the same spot al1 day.
28
THE MACHINE THAT CHANGED THE WORLD
Then, around 1908, wl~enFord finally achieved perfect part
interchangeability, he decided that the assembler would perform
only a single task and move from vehicle to vehicle around the
assembly hall. By August ol1913, just belore the moving assembly
line \vas introduced, the task cycle for the average Ford assembler
liad been reduced from 514 to 2.3 minutes.
Naturally, this reduction spurred a remarkable increase in
productivity, partly because complete familiarity with a single
task meant the worker could per-form it faster, but also because
al1 filing and adjusting of parts had by now been eliminated.
Workers simply popped on parts that fitted every time.
Ford's innovations must have meant huge savings over earlier
production techniques, which required workers to tile and fit each
imperfect part. Unfortunately, the significance of this giant leap
toward mass production went largely unappreciated, so we have
no accurate estimates of the amount of effort-and money-that
the minute division of labor and perfect interchangeability saved.
Wc do know that it was substantial, probably much greater than
the savings Ford realized in the next step, the introduction in
1913 of the continuous-flow assembly line.
Ford soon recognized the problem with moving the worker
from assembly stand to assembly stand: Walking, even if only for
a yard or two, took tinie, and jam-ups frequently resulted as faster
workers overtook the slower workers in front of them. Ford's
stroke of genius in the spring of 1913, at his new Highland Park
plant in Detroit, was the introduction of the moving assembly
line, which brought the car past the stationary xvorker. This
innovation cut cycle time from 2.3 minutes to 1.19 minutes; the
difference lajf in the time saved in the worker's standing still
ratlier than walking and in the faster work pace, which the
moving line could enforce.
With this highly visible change, people finally began to pay
attention, so we have well-documented accounts of the manufacturing effort this innovation saved. Journalists Horace Arnold and
Fay Faurote, for example, writing in Engineering Muguzine in 1915,
compared the number of items assembled by the same number of
workers using stationary and moving-assembly techniques and
gave the world a vivid and dramatic picture of what Ford had
wrought (see Figure 2.1)
Productivity improvements of this magnitude caught the attention and sparked the imagination of other auto assemblers.
Ford, his competitors soon realized, had made a remarkable
29
THE RISE AND FALL OFMASS PRODUCTlON
Craft Production versus Mass Production in the Assembly Hall:
1913 versus 1914
Mniutes of
Elforl lo
Assembfe:
Engine
Magneto
Axle
Major Components
into a Complete
Vehicle
Note:
h
Late Craii
Prodocfion,
Fzli 1913
Mass
Pioducfion,
Spring lSi4
226
Percmi
'Icducfion
in EiIori
20
150
26.5
62
75
83
750
93
88
594
5
"Late craíi production" aiready contained rnany of the elements of mass production. in
particular consistently interchangeable parts and a minute division o i labor. The big change
from 1913 to 1914 was the transition írom stationary fo rnoving assernbly.
Saurce: Calculated by !he authors from data given in David A. Hounshell, From the American System
to Mass Producfion, 1800-1932, Baltimore: Johns Hopkins University Press, 1904. pp. 248.
254,255, and 256. Hounshell's data are based on the observations of the journalists Horace
Arnold and Fay Faurote as reported in lheir volurne Ford Methods and ihe Ford Shws, New
York Enyineering Magazine. 1915
discovery. His new technology actually reduced capiial requirements. That's because Ford spent practically nothing on his assembly line-less than $3,500 a t Highland Park6-and it speeded
up production so dramatically that the savings he could realize
from reducing the inventory of parts waiting to be assembled far
exceeded this trivial outlay.
(Ford's moving assembly consisted of two strips of metal
plates-one under the wheels on each side of tlie car-that ran
the length of the factory. At the end of the line, the strips, rnounted
on a belt, rolled under the floor and returned to the beginning.
The device was quite similar to the long rubber belts that now
serve as walkways in some airports. Since Ford needed only the
belt a n d a n electric motor to move it, his cost was minimal.)
Even more striking, Ford's discovery simultaneously reduced
the amount of human effort needed to assemble an automobile.
What's more, the more vehicles Ford produced, the more the cost
per vehicle fell. Even when it was introduced in 1908, Ford's
Model T, with its fully interchangeable parts, cost less than its
rivals. By the time Ford reached peak production volume of 2
million identical vehicles a year in the early 1920s, he had cuc the
real cost to the consumer by an additional two-thirds.7
30
THE MACHlNE THAT CHANGED THE WORLD
To appcal to his target mal-ket of average consumcrs, Ford
had also designed uriprecedented case oi opcration ancl maintainability into Iiis car. He assumed that his buyer would be a iarmer
with a modest tool kit and tlie kinds oS mechanical skills riceclcd
for fixing Sarm macliinery. So the Model T's uwner's manual,
ivhich ivas ivritten in question-and-answer forin, explained in
sixty-four pages how owriers could use simple 1001s Lo soive any
of the 140 problcins likely LO crop up with the car.
For example, ocvners could remove cyliiider-head carbon,
which causes Itnoclring and power loss, Iroin chamber roofs and
piston cro;vns by loosening the lifteen cap screws that held tlic
cylinder head and using a putty knik as a scraper. Simiiarly, a
single paragraph and une diagram told customcrs how LO remove
carboii deposits from their car's vali~cstvitli thc Ford Valve Grinding Tool, which carne with the auto.x And, if a part needed
replacement, owners could huy a spare at a Ford dealer and
simply screw or bolt it on. With the Ford Model T, there was no
fitting required.
Ford's competitors were as amazed by this desigiled-in repairability as by the moving assembly line. This con.ibination of
competitive advantages catapulted Ford tu the head OS the world's
motor industry and virtually elirrrinated craft-production companies unable to match its manufacturing economies. (As we pointed
out earlier, however, a few European craft-based producers of
ultra-low-volume luxury cars cvuld ignore tbe juggernaut of mass
production.)
Henry Ford's mass production drove tlie auto industry for
more than half a century aild was eventually adopted in almost
every industrial activity in North America and Europe. Now,
however, those s a n e tecliniques, so ingrained in manufacturing
philosophy, are thwarting the eSSorts OS maily Western compariies
to move ahead to lean production.
What precisely are the characteristics OS mass production as
pioneered by Ford in 1913 and persisting in so many companies
today? Let's take a look.
Work Force
Ford not only perfected the interchangeable part, he perfected the
interchailgeable worker. By 1915, when the assemblv lines at
THE RlSE AND FALL OF MASS PRODUCTION
31
Highlaiid Park were Sully installed and output reached capacity,
assembly workers numbered more than 7,000. Most were recent
arrivals to Detroit, often coming directly from the farm. Many
more were riew to the United States.
A 1915 survey revealed that Highland Park worlters spoke
more than fifty lailguages and that many of them could barely
speak English.' How could this army OS strangers cooperate to
produce a greater volume OS a complex product (the Model T)
an any company had previously imagined-and do it with
consistent accuracy?
The answer !ay in takiilg the idea of the division of labor to
its ultimate extreme. The skilled fitter in Forci's craft-produciion
plant of 1908 had gathered al1 the necessary parts, obtaincd tools
h-om the tool room. repaired them if necessary, performed the
complex fitting and assembly job for the entire vehicle, then
checked over his work before sending the completed vehicle to the
shipping department.
In stark contrast, the assembler on Ford's mass-production
line had only one task-to put two nuts on two bolts or perhaps
to attach one wheel to each car. He didn't order parts, procure his
tools, repair his equipment, inspect Sor quality, or even understand what the ivorlters on ejther sitie of him were doing. Kather,
he kept his head down and thought about other things. The Sact
that he might not even speak the same language as his fellow
assemblers or the foreman was irrelevant to the success of Ford's
system. (Our use of "he," "him," and "his" is deliberate; until
World War 11, workers in auto factories in the United States and
Europe were exclusively male.)
Someone, of course, did have to think about how al1 the parts
canle togetlier and just what each assembler should do. This was
the task for a ncwly created professional, the industrial engineer.
Similarly, someone had to arrange for the delivery of parts to the
line, usually a production engineer who designed conveyor belts
or chutes to do the job. Housecleaning workers were sent around
periodically to clean up work areas, and skilled repairmen circulated to refurbish the assemblers' tools. Yet another specialist
checked quality. Work that was not done properly was not discovered until the end of the assernbly line, where ariother group of
workers was called into play-the rework men, who retained
many of the fitters' skills.
With this separation of labor, the assembler required only a
few minutes of training. Míoreover, he was relentlessly disciplined
32
THE MACHINE THAT CHANGED THE WORLD
by the pace of [he linc, which speeded up the slo\v and slowed
down the speedy. The Soreman-formerly the head os a whole
arca of the factory with wide-ranging duties and responsibilities,
but no\v reduced to a semiskilled checlter-could spot immediately any slacking off or failure t« perform the assigned task. As a
result, the workers on the line were as replaceable as the parts on
the car.
In this atmosphere, Ford touk it as a given that his worlcers
wouldn't volunteer any insormation on operating conditions-for
example, that a tool was malfunctioiling-much less suggest w q s
to improve the process. These functions Sell respectively to the
foreman and the industrial crigineer, who reported their findings
and suggestions to higher levels «S managemcnt Sor action. So
were born the battalions of narrolvly skilled indirect 'ivorkersthe repairman, tlie quality inspector, the housekeeper, and the
rework specialist, in addition to the foreman and the industrial
engineer. These workers hardly existed in eral1 production. Indeed, Faurote and Arnold never thought to look for them when
preparing the productivity figures shown in Figure 2.1.'"These
figures count only the dircci workers standiilg on the assembly
line. However, indirect workers became ever more prominent in
Fordist, mass-production lactories as the introduction of aiitomation over the years gradually reduced the need for assemblers.
Ford was dividing labor not only in thc factory, birt also in
the engineering shop. Industrial engineers took their places next
to the manufacturing engineers who designed the critica1 production machinery. They \vere joined by product engineers, who
designed and engineered the car itself. But these specialties were
only the beginning.
Some industrial engineers specialized in assembly operations, others in the operation of the dedicated machines making
individual parts. Come manufacturing engineers specialized in
the design of assembly hardware, others designed the specific
machines for each special part. Some product engineers specialized in engines, others in bodies, and still others in suspensions
or electrical systems.
These original "knowledge wor1cers"-individuds who manipulated ideas and information but rarely touched al1 actual car
or even entered the factory-replaced the skilled machine-shop
owners and the old-Sashioned factory foremen ol the earlier craft
era. Those worker-managers had done it all-contracted with the
assembler, designed the part, developed a machine to make it,
THE RlSE AND FALL OFMASS PRODUCTION
33
and, in many cases, supervised the operation of the machine in
the workshop. The fundamental mission of these new specialists,
by contrast, ' i a s lo design tasks, parts. and tools that could be
handled by the unskilled workers ivho made up the hulk of ihe
new motor-veliicle iiidustry work Sorce.
In this new system, the shop-floor worker had no career path,
except perhaps to foreman. Bui. tlie newly emerging professional
engineers had a direct climb up the career ladder. Unlike tlie
skilled craftsman, howevcr, their career paths didn't lead toward
ownership of a business. Nor did they lie within a single company,
as Ford probably hoped. Rather, they would advance wiihin their
profession-Srom yourig engineer-trainee to senior engineer, who,
by now possessing the entire body oF knowledge oi ihe profession,
was in charge of coordinating engineers at lower le'i~els.
Reaching the pinnacle of the engineering profession often
meant hopping from company to company over the course of
one's working lile. As time went on and engineering branched into
more and more subspecialties, these engineering professionals
found they had more and more to say to their subspecialists and
less and less to say to engineers with other cxpertise. As cars and
trucks became ever more complicated, this minute division of
labor wilhin engineering would result in massive dystiinctions,
which we'll look at in Chapter 5.
Henry Ford was still very much an assembler when he opened
EIighland Park. He bought his engines and chassis from ihe Dodge
brothers, tlieii addcd a hosr uf items ordered from other firms to
make a complete vehicle. By 1915, however, Ford had taken al1
these Eunctions in-house and \\las well 0x1 his Lvay to achieving
complete vertical integration (that is, making everything coqnected with his cars from the basic raw materials on up). This
development reached its logical conclusion in the Rouge complex
in Detroit. which opened in 1931. Ford pursued vertical integration partly because he had perfected mass-production iechniques
before his suppliers had and could achieve substantial cost savi n g ~by doing everything himself. But he also llad some other
reasons: For one, his peculiar character caused him to profoundlv
distrust everyone but himself.
34
THE MACHINE THAT CHANGED THE WORLD
However, his mosi important reason for bringing everythiilg
in-house was the fact that he needed parts with closer tolerantes
and on tighter delivery schedules than anyone had previously
imagined. ReIying on arm's-length purchases in the open marketplace, he figured, would be fraught with difficulties. So he decided
to replace the mechanism of the market with the "visible hand"
of organizational coordination.
Alfred Chandler, a professor at the 1-Iarvard Business School,
coined the term "visible hand" in 1977. In his book of the same
title, he attempted Lo provide a defense for the modern large
firm." Proponenis o i Aclam Smith's "invisible hand" theory
(which argued ihat if everyone pursued his or her owii selfinterest, the free market would of itself produce the best outcome
for society as a whole) were disturbed by the rise in the twentieth
century of the vertically integrated modern corporation. In tlieir
view, vertical integration interfercd with free-market forces.
Chandler argued that a visible hand was critica1 if modern corporations ivere to introduce necessary predictability into their
operations.
Chandler used the term simply to mean obtaining needed
raw materials, services, and so forth fi-om interna1 operating
divisions coordinated by senior executives at corporate headquarters. The invisible hand, by contrast, meant buying necessary
parts and services from independent firms with no financia1 or
other relationship to the buyer. Transactions would be based on
price, delivery time, and quality, with no expectation of any longterm or continuing relationship between the buyer and the seller.
The problem, as we will see, was that total vertical integration
introduced bureaucracy on such a vast scalc that it brought its
own problems, with no obvious solutions.
The scale oí'production possible-and necessary-with Ford's
system Ied to a second organizational difficulty, this time caused
by shipping problems and trade barriers. Ford wanted to produce
the entire car in one place and sell it to the whole world. But the
shipping systems of the day were unable to transport huge volumes of hnished automobiles economically without damaging
them.
Also, government policies, tlien as now, often imposed trade
barriers on finished units. So Ford decided to design, engineer,
and produce his parts centrally in Detroit. The cars, however,
would be assembled in remote locations. By 1926, FOI-d automo-
THE RlSE AND FALL OF MASS PRODUCTION
biles were assembled in more than thirty-six cities in the United
States and in nineteen foreign countries.'"
It wasn't long before this soluticin created yet another problem: One standard produci just wasn't suited Lo al1 world markets. For example, to hmericans, Ford's Model T secmed lilte a
small car, particularly after the East Texas oil discoveries pushed
gasoline prices down and made longer travel by car econornically
feasible. However, in England and in other Europea11 countries,
with their croivded cities and narrow roads, ¡he Model T seemed
much larger. In addition, when tbe Europeans failed to find any
oil at home, they began to tax gasoline heavily in the 1920s to
reduce imports. The Europeans soon began to clamor for a
smaller car than Ford wanted to supply.
Morcover, massive direct iiii~estmentin hreigii countries created resentment of Ford's dominance of local industry. In England, for example, where Ford had become the leading auto
manufacturer by 1915, his pacifism in World War 1 \vas roundly
denounced, and the company's local English managers finally
convinced Detroit io sell a large minority stake in the hiisiness to
Englishmen to diffuse hostility. Ford encountered barriers i n
Germany aiid France as well after World War 1, as tariffs were
steadily raised on parts and complete vehicles. As a result, by the
early 1930s, Ford had established ~ h r e efully integrated manufacturing systems in England, Germany, and France. Thcse companies produced special products for national tastes and were run
by native managers who tried to minimize meddling fi-om Detroit.
The key to interchangeab!e parts, as \ve saw, lav in designing new
tools that could cut hardened metal and stamp sheet steel with
absolute precision. But the key to ii~aperzsiteintcrchangeable
parts would be found in tools that could do this job at high
volume with low or no set-up costs between pieces. That is, for a
machine to do something to a piece of metal, someone must put
the metal in the machine, then someone may need to adjust the
machine. In the craft-production system-wliere ri single machinc
could do many tasks but required lots of adjustment-this was
the skillcd machinist's job.
THE MACHlNE THAT CHANGED THE WORLD
Ford dramatically reduced set-up time by making machines
that could do only one task at a time. Then his engineers perfected
simple jigs and fixtures for holding the work piece in this dedicated machine. The unskilled workers could simply snap the piece
in place and push a button or pul1 a lever lar the machine to
perform the required task. This meant the machine could be
loaded and unloaded by an employee with five minutes' training.
(Indeed, loading Ford's machines was exactly like assembling
parts in the assembly line: The parts would fit only one way, and
the worlrer just popped them on.)
In addition, because Ford made only one product, he could
place his machines in a sequence so that each manufacturing step
led immediately to the next. Many visitors to Highland Park felt
that Ford's factory was really one vast machine with each production step tightly linked to the next. Because set-up times were
reduced from minutes-or cven hours-to seconds, Ford could get
much higher volume from the same number of machines. Even
more important, the engineers also found a way to machine many
parts at once. The only penalty with this system was inflexibility.
Changing these dedicateci machines to do a new task was timeconsuming and expensive.
Ford's engine-block milling machine is a good example of his
new system. In almost every auto engine, then and now, the top of
the engine block is mated to thc bottom of the cylinder head to
form a complete engine. To maintain compression in the cylinders, the fit between block and head must be absolutely flush. So
the top of the block and the bottom of the cylinder tiead has to be
milled with a grinding tool.
At Henry Leland's CadilIac plant in Detroit (where, incidentally, consistent interchangeability for al1 the parts in a motor
vehicle was achieved for the first time in 1906), a worker would
load each block in a milling machine, then carefully mil1 it to
specification. The worker would repeat the process for the cylinder heads, which were loaded one at a time into the same milling
machine.
In this way, the parts were interchangeable, the fit between
block and head was flush, and the milling machine could work on
a wide variety of parts. But this process had a down side: the rime
and effort-and therefore expense-it took for the skilled machinist who operated the machine.
At Highland Park in 1913; Ford introduced two dedicated
machines, one Sor milling blocks and the other for rililling heads-
THE RlSE ANO FALL OF MASS PRODUCTlON
37
not just one at a time, but fifteen at a time for blocks and thirty at
a time for heads. Even more significant. a fixture on both machines allowed unskilled workers to snap the blocks and heads in
place on a side tray, while the previous lot was being milled. The
worker then pushed the whole tray into the miller, and the process
proceeded automatically. Now the entire ski11 in milling was
embodied in the machine, and the cost of the process plummeted.
Ford's tools were highly accurate and in many cases automated os nearly so, but they were also dedicated to producing a
single item, in some cases to an absurd degree. For example, Ford
purchased stamping presses, used to make sheet-steel paris, with
die spaces large enough to handle only a specific part. When the
factory needed a larger part because of a specification change or,
in 1927, for the completely redesigned Model A, Ford often discarded the machinery along with the old part or model.
Ford's original mass-produccd product, the Model T, carne in nine
body styles-including a two-seat roadster, a four-seat touring
car, a four-seat covered sedan, and a two-seat truck with a cargo
box in the rear. Howvever, al1 rode on the same chassis, which
contained al1 the mechanical parts. In 1923, the peak year of
Mndel T production, Ford produced 2.1 million Model T chassis.
a figure that would prove to be the high-water mark Sor standardized mass production (although the VW Beetle later equaled it).
Tlie success of his automobiles was based first and foremost
on low prices, which kept falling. Ford dropped his prices steadily
from the day the Model T was introduced. Some of the reduction
had to do with shifts in general consumer prices-before gowrnments tried to stabilize the economy, consumer prices went down
as well as up-but mostly it was a matter of growing volume
permitting lower costs that, in turn, generated higher voiume. At
the end of its run in 1927, however, Ford was facing falling
demand for the Model T and was undoubtedly selling below cost.
(Demand fell, because General Motors was offering a more modern product for only a little more money. Moreover, a one-yearold GM automobile was less cxpensive than a new Ford.)
The Ford car's amazing popularity also stemmed Srom iis
durability OS design and materials and, as noted, from the fact
38
THE M C H I N E THAT CHANGED THE WORLD
that the average user could easily repair i t . Concerns thai buycrs
rank highest today hardly existed in Ford's world.
For example, fits and finishes-or the cosrnetic aspects ol a
car, such as the Sendcr panels coming together without gaps, a
lack of dribbles in ihe paint, 01- the doors rnaking a satisfying
clrinlting sound when slammecl-wcren't a c«ncern for Ford's
custoniers. Tlie Mciclel T liad no exterior sheci i~ietaicxccpi. the
hood; the paint was so crude thai you would hai.dly Iiave noticed
drihbles; anci several ol'the body styles had no dool-s at all.
4s ior breakdowiis or problems in daily use-engi~ies that
stumble, say, or mysicrious clectrical difhculties, srrch as the
"clicck cnginc" signal that conies on periodically in some carsthese, too, didn't hotber Ford's buyers. IL thc R/Iodel T cngine
stumbled. thcy simply looked Sor the cause in the question-alidanswei- booklet the company provided and iixed the problem. For
example, tliey might drain the gas tank and pour the Suel back
ihrough a chamois to strain any water oui. Thc bottorn line: If a
part ciidn't fit pruperly or ivas installed slightly out of tolcrance,
the owner \vas expected to iix i i . And, since al1 cars broke clown
frequentlv, ease of rcpair was key.
At Highland Parlc, Ford rarely inspected finished automobiles.
No one ran an enginc until the cal- was rcady to drive away Srom
the c ~ i dof tlie assei~iblyline, and 110 Model T was ever road-iested.
Nonetheless, dcspiie a manufacturing systeiu that probably
did riot deliver very liigh qiiality in our modern sense, Ford was
able to dominate what soon became the world's largest industry
by l>ccorningthe first to inasier the principies olrnass production.
It wasn't until fifty years later ihat plants organizcci oii leanproductiori pririciples could deliier ricai--pcrkct qualiiy witho~it
exteiisii~ecnd-of-the-linc inspeciions and large amounts of rcwork.
1
THE RlSE AND FALL OF MASS PROUCTION
39
At these facilities, Ford continued his obsession with a single
product-the Model A a t the Rouge, the Model Y at Dagenham,
and the Ford V8 in Germany. He also added a steel mil1 and a
glass factory to the metal-forming and -cuttingaclivitics that took
place at Highland Park. Al1 the necessary raw materials now came
in one gate, while finished cars went out the otlier gate. Ford
had succeeded in completely eliminating the need for outside
assistance.
He even added raw materials and transport to the visible
hand-through a wholly owned rubber plantation in Brazil, iron
mines in Mimitisota, Ford ships to carry iron ore and coa1 ihrough
the Great Laltes to tlie Rouge, aiid a railroad to connect Ford
production facilities in the Detroit region.
In the end, Ford attempted to mass-produce everything-from
food (through tractor manufacture and a soybean extraction
plant) to air transportation (by means o l the Ford TriMotor, wliich
was supposed to reduce the price oí commercial air iralfic, and
the Fo1-d "Flying Fliver," which \vas intended as the airborne
equivalent of tlie Model T). Ford's idea was that by making
everything, from food to tractors to airplanes, in a standardized
form at high volume, lie could dramatically reduce the cost oí
producls and make the masses rich. He financed al1 his projects
internally, Sor Ford loathed banks and outside investors and was
determined to maintain total control of his company.
Eventually, these steps beyond Highland Park al1 came to
naught, partly because the synergy among industries, which industrialists repeatedly seek and seldom find, was never there, but
also because Ford himselí had absolutely no idea how to organize
a global business except by centralizing al1 decision-making in
the one person at the top-himself. This concept was unworkable
even when Ford was in his prime, and it nearly drove the company
under when his mental powers declined in the 1930s.
THE LOGIEIL LIMKS OF MLSS PROOUETMN: THE ROüGi
Ti-uc ~iiassproduction began wiih Higliland í'ark, but tiic erid
wasn't yet in sight. Ford hclicveci tliai the last piece in the puzzle
was to apply a "visible hand" to every step iii production, from
raxv materials to finished vehicle. This Iic attemptcd to do at tlie
Rougc complcx, which he openeci ricar- Detroit in 1927. Siriallerscale duplicates oí'the Ro~igewerc opcned at Dagenhani, Eilgland,
and Coiogne, Gci-rnany, in 1931 .
SLOAN AS A NECESSARY GOMPLEMENT TO FORO
Alíred Sloan at General Motors already had a better idea in the
early 1920s when he was called in to straighten out the messes
that William Durant, General Motors' mercurial ioulider, had
made. Durant was the classic empire-building financier. He had
absolutely no idea how to manage anything once he bought i t . He
40
THE MACHINE THAT CHANGED THE WORLD
therefore wound up with a dozen car companies, each managed
separately with a high degree of product overlap. Because he had
no way to know xvhat was going on in these companics, beyond
quarterly profit-and-loss statements, he was repeatedly surprised
to discover that too many cars were being manulactured for
market conditioiis or that not enough raw materials were available to sustain production. A burst of oi~erproductionheading
into the economic slump of 1920 finallv did him in; his bankers
insisted that someone with management skills take the helm. So
Pierre du Pont, chairman of E. 1. du Pont, became chairman of
General Motors and, in turn, made Sloan GM's president.
An MIT graduate (he contributed a block of his GM earnings
to found thc Sloan School of Managcment at MIT after World War
II), Sloan gained control in the early 1900s of the Hyatt Roller
Bearing Company, a firm purchased by Billy Durant around 1915.
He was vice-president of GM when Durant was ousted; he gained
the presidency on the basis of a memo he wrote in 1919 on how to
run a multidivisional company.
Sloan quickly saw that GM had two critica1 problems to solve
if it was going to succeed at mass production and oust Ford as the
industry leader: The company had to manage proíessionally the
enormous enterprises that the slew production techniques had
both necessitated and made possible, and it had to elaborate on
Ford's basic product so i t would serve, as Sloan put it, "every
purse and purpose."
Ford Motor Company, of course, didn't suffer from GM's
product overlap problem, because Ford produced only one product. It did, liowever, have al1 the organizational problems, but
Henry Ford refused to acknowledge them. He succeeded with
mass production in the factory, but he could never devise the
organization and management system he needed to manage effectively the total system o€ factories, engineering operations, and
marketing systems that mass production calleci for. Sloan would
make the system Ford had pioneered complete, and it is this
complete systcm to which the term tnussproduction applies today.
Sloan swiftly found a solution for each of GM's difficulties. To
resolve the management problem he created decentralized divisions managed objectively "by the numbers" from a small corporate headquarters. That is, Sloan and the other senior executives
oversaw each of the company's separate profit centers-the five
car divisions and the divisions making components such as generators (Delco), steering gears (Saginaw), and carburetors (Ro-
THE RlSE AND FALL OF MASS PRODUCTION
41
chester). Sloan and his executive group demanded detailed reports a t frequent intervals on sales, market share, inventories,
and profit and loss and reviewed capital budgets when the divisions required funds from the central corporate coffer.
Sloan thought it both unnecessary and inappropriate for
senior managers a t the corporate level to know much about the
details of operating each division. If the numbers showed ihat
performance was poor, it \vas time to change the general manager. General managers showing consistently good numbers
were ca~ldidatesfor promotion to the vice-presidential leve1 at
headquarters.
To satisfy the broad niarltet General Motors wanted to serve,
Sloan developed a five-model product range that ran incrementally from cheap to expensive, from Chevrolet to Cadillac. It
would, Sloan reasoned, fully accommodate potential buyers of
every income throughout their lives.
Sloan had worked out this strategic solution to the company's
problems by about 1925, although he only codified it for the world
outside General Motors when he got around to writing his memoirs as he approached ninety in the 1960s.')
He also worked out solutions to t\vo other major problems
confronting the company. Through his links with DuPont and the
Morgan Bank, he developed stable sources of outside funding,
which would be available when needed.
Also, his idea of decentralized divisions domestically worked
just as well in organizing and managing GM's foreign subsidiaries. Manufacturing and sales operations in Germany, Britain,
and many other countries hecame self-reliant companies managed by the numbers from Detroii. The arrangement demanded
very little management time or direct supervision.
It isn't giving Sloan too much credit to say that his basic
management ideas solved the last pressing problems inhibiting
the spread of mass production. New professions of financia1 managers and marketing specialists were created to complement the
engineering profcssions, so that every functional asea of the firm
noxv had its dedicated experts. The division of professional labor
was complete.
Sloan's innovative thinking also seemed to resolve the conflict
between the need for standardization to cut manufacturing costs
and che model diversity required by the huge range of consumer
demand. He achieved both goals by standardizing many mechanical items, such as pumps and generators, across the company's
esisting body dcsigns to sustain consumer inierest.
Cloan's innoir~tions~,vci-e
a rcirolution in marketing a
agcrricnt fur the auto industry. However, they did no
ciiange tlie idea, institui.iona1ized firsi by Henry Ford,
ivorkers on thc shop floor \.vere simply interchangeable
thc protluclioii systcm. Sol on the sliop floor matters we
bad io much worsc.
fi>rd bimself was happ), enough with ttlc Iiigli leates
over his labor philosophy and practices encouraged. None
he realizcd that once rhe continuous-flow system was f
place at Highland Park in 1914, his company's elficiency
much higher than its riirals ¡liar he could afford simulta
slash priccs. These actions permitted hini to portray himse
paieriialistic employer (and avoid unions), ivhile he dro
crah-hascd competitors to the wall.
The trouble with the higher- wage, as it turned out, wa
r equity and Eairness.
UCTION: AMERICA IN 1955
y practica, add Sloan's marketing and manageand mi* in organized labor's new role in connments and work tasks, and you have mass
final mature form. For decades this system
victory to victory. The U.S. car companies domild automotive industry, a ~ i dthe U.S. markct acpracticaIly every other industry adopted similar
lly leaving a few craft firms in low-volurrie nichcs.
thcir conditions oi emplqmcni rapidiy carne to seem
less bearablc.
Furtliermorc, the auto iiiarket turricd out lo be even
Depressioii the conditioiis for a successful union movemen
auto industry were fully in place.
Icadership iully accepted both tlic role o{ m;inagement a
inhercnt nature ol'work in an asscnibly-line factory. Not s
ingly, then, ivlien tlie United Auto Workers finally signed a
nlcnls wilh whal had become the Big Tliree iri the late 1930
niail? issues wcrr seriiority and Job rights; the movemen
called job-cont1.01uiiionisiri."
retired after thirtv-four years as either presiden1
of General Motors.
t o[all sales, and six modcls accounted Sor 80 pcrcent
]d. ~ 1 vestiges
1
o[ craft production, once the way of
were now gone in the United States.
fleeting, however, as the then mighty U.S. auto
ow ]carned. Ironically, 1955 was also the year that
slide began, as Figures 2.2 and 2.3 show. The &are
imed by imports begari its steady rise. Their e a r l ~
44
THE MACHINE THAT CHANGED THE WORLD
THE RlSE AND FALL OFMASS PRODUCTION
FIGURE 2.2
Shares of World Miofor Vehicle Production by Region, 1955-1989
Companies, 1955-1989
100,
Year
Note:
This figure includes ail vehicles produced within the three major regions, by al1 companies
operaling in those regions. In addition. it groups the prodiicticn of the newly industrializing
countries and of the rest of the world.
NA = North America: United States and Ganada
E
= Western Europe, inciuding Scandinavia
J
= Japan
NIC = Newiy induslriaiizing countries principally Korea, Brazit, and Mexico
ROW = Rest o1 the world. inctuding the Soviet Union, Eastern Europe, and China
Year
Nofe:
These shares inciode vehicles imported by the American-owned iirms from their wholly
owned and joint-venture factories abroad. They do not include "captive" imports purchaseci
from independent foreign firms.
Source: 1955-1981 from Autornotive News Avarket Dala Book, based on vehicie registrations.
1982-1989 from Ward's Autornobve Reporls, based on vehicie Sales.
Source: Calculaled by the authors from Automotive News Marke! Dala Book, 1990 edition. p. 3.
THE OlFfUSlON UF MASS PRURUCTION
A major reason ¡he Big Three American firms were Iosing their
competitive advantage was that by 1955 mass production had
become commonplace in countries across the world. Many people,
in fact, had expected thc American lead to narrow much earlier,
in the years immediately after World War 1. Even before the war,
a steady stream of pilgrims, including André Citroen, Louis Renault, Giovanni Agnclli (o[ Fiat), Herbert Austin, and William
Morris (of Morris and MG in England), had visited Highland Park.
Henry Ford was remarkably open in discussing his techniques
with them, and, in the 19305, he directly demonstratcd every
aspect of mass production in Europe mith his Dagenham aiid
Cologne factories.
The basic ideas underlying mass production had, thcrzfore,
been freely available in Europe ior years belore rlie onsei ol Wor-ld
War 11. However, the ecoriomic chaos and narrow nationalism
existing therc during the 1920s arid early 1930s, along with a
strong attachment to the craft-production traditions, prcvented
thern from spreading very far. A t the end of the 1930s, Volkswagen
and Fiat began ambitious plaris los mass production at Wolfsburg
and Mirafiori, but World War 11 soon put civilian production oii
hold.
So, it wasn't until thc 1950s, more theri thirty years after
Henry Ford pioneered h i g h - ~ i o l ~ ~mass
m e production, that this
technology, unremarkably commonplace in tiie United Siates,
46
JHE MACHINE JHAJ CHANGED JHE WORLD
hlly diffused beyond Ford's native turf. By the late 1950s, Wolfsburg (VW), Flins (Renault), and Mirafiori (Fiat) were producing at
a scale comparable to Detroit's major facilities. Furthermore, a
number of the European craft-production iirn~s,led by DaimlerBenz (Mercedes), also made the transition to mass production.
All these companies were offering products tllat were distinctly different from the standai-d-size car and pickup truck
lavored by the U.S. manufacturers. 111the early days, the Europeans specialized in two types of cars that the Americans didn't
offer: compact, economy cars, exemplified by the VW Beetlc, and
sporty, fu11-to-drivecars, such as the MG. Laier, in the 1970s, they
redefined the luxury car as a somewhat smaller vehicle with
higher technology and more sporting road manners (the 3,500pound, hel-injected, independently suspended, unibody Mercedes
versus the 5,000-pound, carbureted, straight-axle, body-on-chassis Cadillac). (The unibody car weighs less for a given size of
passenger compartment than a body-on-chassis car. Though it
has the advantages of greater rigidity and thus less of a tendency
to rattle, il also costs more to engineer.)
Combined with Europe's lower wages, these product variations were their competitive opening into world export markets.
And, like the Americans before them, Europeans racked up success after success in foreign markets over a period of twenty-live
years, from the early 1950s into the 1970s.
They also concenirated-as Detroit did not during this timeon introducing new product features. European innovations in the
1960s and 1970s included front-wheel drive, disc brakes, fue1
injection, unitized bodies, five-speed transmissions, and engines
with high power-to-weight ratios. (Unitized bodies have no li-ame
of steel beams under the car. Instead, like a tin can, the surface
sheet metal holds the car together.) The Americans, by contrasi,
were the leaders in comfort features-air conditioning, power
steering, stereos, automatic transmissions, and massive (and very
smooth) engines.
History could have gone the Americans' way if fue1 prices had
continued to fall-as they did for a generation, up until 1973and i l Americans had continued to demand cars that isolated
them from their driving environment. However, energy prices
soared and younger Americans, particularly those with money,
wanted something fun to drive. Detroit's problem \vas that its
"hang-on" features, such as air conditioning and stereos, could
easily be added to existing European cars. But it would take a
JHE RlSE AND FALL OF MASS PRODUCJlON
47
total redesign of the American vehicles and new production tools
to introduce more space-efficient bodies, more rcsponsive suspensions, and more fuel-efficient engines.
However, as became apparent in the late 1980s and as \ve \vil1
show in the Follo~vingchapters, rhe Europea11produciion systems
were nothing more than copies of Detroit's, but with less efficiency and accuracy in the factor>!.
European auto plants experienced in tlle 1950s what rhe
Americans had experieilced in the 1930s. During the early posrwar
years, most European plants emploved large numbers of immigrants-Turks and Yugoslavs iii Germany, Sicilians and other
southern Italians in Italy, and Moroccans and Algerians in
France-in tlie interchangeable assembler jobs.
Some of these people returned home as the posturar European
labor shortage eased. Othcrs, hoxvever, stayed, to be joined by
larger numbers of native workers. Evcntually, just as had happened in the United States, the workers in Turin, Paris, and
Wolfsburg realized that mass-production work was no¡ a way
station to self-employment back home; it was, instead, their life's
work. Suddenly the interchailgeable, dcad-end monotony of massproduction plants began to seem uiibearable. A wave of unrest
followed.
The European rnass-production systems were patched up in
the 1970s by increasing wages and steadily decreasing the weekly
hours of work. European car makers conducted a few marginal
experiments as wcll with worker participation, such as thc one at
Volvo's Kalmar plant, which-in a reviva1 of Henry Ford's assembly hall of 1910-reintroduced craft techniques by giving small
groups of workers responsibility for assemblirig a wholc vehiele.
In addition, the sobering economic conditions after 1973 damped
worker expectations and reduced employment aliernatiipes.
These were only palliativcs, however. In the 19ROs, European
workers continued to lind mass-production work so unrcwarding
that the lirst priority in negotiations continued to be reducing
hours spent in the plant.
This situation of stagnant mass production in both the United
States and Europc might have continued indefinitely if a new
motor industry had not emerged in Japan. The true significance
of this industry was that it \vas not siinply another replicatiun of
he by norv venerable Ainerican approach ¡o mass productioii.
The Japanesc \vere developing an entirely new \\my of making
things, which we cal1 leun prodtiction.
:,-.,
Eiji was not an average engineer, either in abiliiy or ambitiun.
After carclully studying every inch of the m s t Rouge, then 11112
largest and most efficient inanufacturing facility in the \vorld, Eiji
wrote back to headquarter-s that he "thoughi tliere \vere .sonic
possibilities to improve the produciion system.""
But simply copying and improving the Koi~geproved io bc
hard worlr. Back at home in Nagoya, Eiji Toyuda and his production genius, Taiichi Ohno, soon c«ncluded-Sor reasons we \vil1
explain shortly-that mass production could ncver wurk in Japan.
From this tentative begiiining were horn whai T q o t a carne to cal1
he Toyota Production System nnd, ultimately, lean prr~dilctiuii:'
THE RlSE OF LEAN PROWCTION
THE BIRTHPLRGE OF LEAN PRODUCTION
in the spring of 1950, a young Japanese engineer,
Eiji Toyoda, set out on a three-month pilgrimage
to Ford's Rouge plant in Detroit. In fact, the trip
marked a second pilgrimage for the Samily, since
Eiji's uncle, Kiichiro, had visited Ford in 1929.
Since that earlier time much had happeiled to thc Toyoda
family and thc Toyota Motor Company they had founded in 1937.'
(The fouriding family's naiue, Toyoda, means "abundant rice
field" in Japanese, so marketing considerations called los a new
name Sor the íledgling company. Accordingly, in 1936, the company held a puhlic eontest, which drew 27,000 suggestions. "Toyota," which has no meaning in Japanese, was the winner.)
Most os these events had been disastrous lor the company:
They had been thwarted by the military government in their
effort to build passeiiger cars in the 1930s, and had instead made
trucks, largely with craft methods, in the ill-fated war effort.
And, at the end of 1949, a collapse in sales Sorced Toyota to
terminate a large part of the work force, but only after a lengthy
strike that didn't end iintil Kiichiro resigned from the company
to accept responsibility for management failurcs. In thirteen
years of effort, the Toyota Motor Company had, by 1950, produced
2,685 automobiles, compared urith the 7,000 the Rouge was pouring out in a single day.'
This was soon to change.
Toyota is often called the most Japanese oS the Japancse auto
companies, being located in insular Nagoya rather than cosmopolitan Tokyo.' For many years its wol-k force \vas composcd
largely of former agricultui-al workers. in Tokyo, the 6rn1 was
often derided as "a bunch of fariners." Yet toda-, T q o t a is regarded by most industry observers as the most efficient and
highest-quality producer of motor vehicles in tfie world.
The founding Toyoda family succceded tirst in ihe ~ e x t i l e
machinery business during the late nineteciith cenrury by developing superior technical features on its looms. In the late 1930s,
at the government's urging, the company cntered the motor veliicle industry, specializing in trucks for the miliiary. It had barely
one beyond building a few prototype cars with craft methods
elore war broke out and auto produc~ionended. After the wai-,
ota was determiried ro go into full-scale car and cvmmcrcial
k manufaciuring, but it faccd a host of problems.
* The domestic markei was tiny and demancied a rvide range
of i.rehicles-lirxury cars lor government ofíicials, large
trucks to carry goods to market, small trucks for Japan's
small Iarmers, and small cars suitable Los Japan's crowded
cities and high energy prices.
Tbe native Japanese work force, as Toyota and other iirrns
soon learned, was no longer willing to be treate~ias a
variable cost or as iritei-changeable parts. What \+lasmoi-e,
the new labor laws introduced by the America11occupation
THE MACHlNE THAT CHANGED THE WORLD
They then insert the blanks in iilassive stamping presses containing matched uppcr and lower dics. When these dies are pushed
rogcthei- under- thoi~sandsof pounds of pressure, the two-dimensional blanli ~ a k e sthe three-dimensional shape o l a car fender or
a truck door as it moves through a series of presses.
The problem with this seconci method, from Ohno's perspectivc, was the minimum scale required for economical operation.
The massive and expensive Western prcss lines were designed to
operate at about t\velve strokes per minute, three shifts a day, to
rriake a rriilliori or more of a giwn part in a year. Yet, in the early
days, Toyoia's entire prod~ictionwas a few ihousand vehicles a
year.
The dies could be changed so that the same press line could
make many parts, but doiiig so presented major dilficulties. The
dies weighed many tons each, and workers had to align them in
the press with absolute precision. A slight misalignment produced
wrinkled parts. A more serious misaligninent could produce a
nightmare in which the shcet metal melted in the die, necessitating extremely expensive and time-consuming repairs.
To avoid these problems, Detroit, Wolkburg, Flins, and Mirafiori assigneci die changes to specialists. Die clianges were undertalten niethodically and typically required a full day to go from
ihe last part with the old dies to the first acceptable part from the
new dies. As volume in the Western industry soared alter World
War 11, the industry found an even better solution to the diechange problem. Manufacturers found thcy ofteu could "dedicate" a set of presses to a specific part and stamp these parts for
months, or even years, without changing dies.
To Ohno, however, this solution was no solution at all. The
dominant Western practice required hundreds of stamping
presses to make all the parts in car and tmck bodies, while Ohno's
capital budget dictated that practically the entire car be stamped
Srom a few press lines.
His idea was to deveiop simple die-change techniques and to
change dies frequently-every two to three hours versus two to
three months-using rollers to move dies in and out of position
and simple adjustment mechanisms.'Because the new techniques
were easy to master and production workers were idle during the
die changes, Ohno hit upon the idea of letting the production
workers perform the die changes a s \riell.
By purchasing a few used American presses and endlessly
experimenting from the late 1940s onward, Ohno eventually per-
JHE RlSE OFLEAN PRODUCTION
feeted his technique for quick changes. By the late 19.5Os, he had
educed the time required to change dies from a day to an
astonishing three minutes and eliminated the need for iiie-change
cialists. In the process, he made an unexpecied discovery-it
tually cost less per part to make small batches of siampings
an lo run off enormous lots.
There were two reasons lor rhis phcriomenon. Making s n ~ a l l
tches eliminated the carrying cost ol the Iiuge invcntories of
nished parts that mass-production systems requireci. Evcn more
portant, making only a fe\v parts beforc assembling ihcni into
ar caused stamping mistakes to show up almost instantly.
The consequences of this latter discovery were enormous. Ii
ade those in the stamping shop much more concerned about
quality, and it eliminated the wastc ol large riumbcrs ofdefective
parts-which had to be repaired a i great expcnse, or cuen discarded-that were discovered only long after manufacture. But to
make this system work a t all-a systcm that ideally produced tivo
hours or less of iliventory-Ohno needed both an extremely skillcd ,''
anda highly motivated work force.
If workers failed to anticipate problems belore tliev occurred
and didn't take the initiative to devise solutions, the work of ilie
whole factory could easily come to a halt. Holding back knowledge and effort-repeatedly noted by industrial sociologists as a
salient feature o l al1 mass-production systems-wouid swiitly
lead to disaster in Ohno's factory.
As it happened, Ohno's work Sorce acted to solve this prohlem for
him in the late 1940s. Because ol macroeconomic problems in
Japan-the occupying Americans had decided to stamp out inflation through eredit restrictions, but overdid it and caused a
depression instead-Toyota found its nascent car business in a
deep slump and was rapidly exhausting loans from its banlters.
The founding family, led by Presiden1 Kiichiro Toyoda, proposed-as a solution to the crisis-firing a quarter of the work
force. However, the company quickly found itself in the midst of a
revolt that ultimately led to its workers occupying the factory.
Moreover, the company's union was in a strong position to win
the strike. In 1946, when the Japanese government, under Ameri-
54
55
THE MACHlNE THAT CHANGED THE WORLD
can prompting, strengthened the rights of unions, including ma
agement, and theri imposed severe restrictions on the ability
cornpany owners to fire xvorkers, the balance of power had shiSted
to the employees.
After protracted negotiations, the family and the union
worked out a compromise that today remains the lí.)rmula lor
labor relations in the Japanese auto industry. A quarter- uf the
work force was terminated as originally pmposed. But Kiichiro
Toyoda resigned as presidcnt to take responsibility Sor thc company's failure, arid the remaining emplqfees received two guarantees. One \vas foi- lifeti~ncemployment; the other was for pay
steeply gl-aded by seniority rather than by speciíic job lunction
and tied to company profitability through bonus paymeiiis.
ln short, they became members of the Toyota community,
with a full set of rights, including the guarantee of lifctime
employment and access to Toyota lacilities (housing, reciz:ation,
and so forth), that went far beyond what most unions hacl been
able to negotiate for rnass-production employees in the Wcst. In
return, thc company expected that most employees would ren~ain
with Toyota for their working lives.
This was a reasonable expeciation because other Japanese
cornpanies adopted seniority-based wages at the same time, and
workers would suffer a large loss of earnings iS they started uver
at the bottom of the seniority ladder with anotlier company. The
wage progression was quite steep. A forty-year-old \vol-ker doing
a given job received much higher pay than a twenty-tive-year-old
doing the same job. If the forty-yeur-old quit and went to work Sor
another employer, he would start with a zero seniority wage tha
was below that of even the twenty-hve-year-old.
The employees also agreed to be flexible in work assign
and active in promoting the interests olthe conipany by initiating
improvements rather than merely responding to problerns. In
effect, the company officials said: "If we are going to take you on
for lile, you have to do your part by doing the iobs that need
doing," a bargain to which the unions agreed.
Baek at the factory, Taiichi Ohno realized the implications of
this historic settlement: The work force was now as much a shortterm fixed cost as the company's machinery, and, in the long
term, the workers were an even more significant fixed cost. After
all, old machinery could be depreciated and scrapped, but Toyota
needed to get the most out of its huntan resources over a fortyyear period-that is, from tlie time new workers entered the
mpany, which in Japan is gciierally bctweeri tlic agcs ol' cigtin and twenty-twu, until tliey reaclied rctircrncnt at age siuty.
se tu continuously cntiarlcc thc ii:or-liers' sliills ;ind
efit of tlicir knoivlctlgc and cxpcrieiicc as well as
AN PRODUCTIDN: FlNRt ASSEMBLY PLRNT
o's rethinking of iirial assembly shaws just ho\v ihis nctv
~ > a c hto hurnan rcsocii-ccs paid enurrrio~isdividcnds SCII. Toythat Foríl's s ~ ~ s t e rassumed
n
thai asse~iibly-line
erfosm one or i~\.osimple tasks, repetiti\.ely aiitl,
int. The foreman did out pci.li>rni
sembly tasks himself but insiead ensured that thc linc worltci-s
llowed orders. These orders or instructions were devised hy t11c
dustrial engirieer, \vho was also responsible lor coniing 11p with
Special repairmcn repaircd iools. Housekeepei.~pcriodically
aned the work area. Special inspectors checked cluality, and
ed, was i-ectihed in a rewoi.k asea
final category of workei-, ilie utility
completed the division of labor. Silice cven higli \sragcs \vere
le to prevent double-digit absenteeism ir1 ii-iost ninss-producnies ni:eded a large grorip of utility
hose employees wbo dirln'i show up
Managers a t headquarters generally g r ~ d e dfactor? managcment on two criteria-yield and quality. Yield was the number o l ,*'
cars actually produced iii relation to ihe scheduled numher, anci
qualiiy was out-the-door quality, alter vehicles liad defcctive par-ts
repaired. Factory managers knew that falling hclo~vtlie nssignctl
production target spelled big irouble, and that mistakes could, il'
necessary, be fixed iri the rcwork arca, after the end ol' ~ l i ciine bui
befoi-e the cars reached the quality checker from hea~lquai-ters
stationed a t the shipping dock. Therefore, it \vas crucial noi to
stop the line unless absolutely necessary. Letting cars go ori down
the line wiih a misaligned part ivas perlectly okay, because thii
type of defect could be rectified in the rework area, bui minutes and cars lost to a line stoppagc coulcl only be rnade up
with expensive overtime at the end of the shift. Thus ivns horn
58
THE MACHlNE THAT CHANGED THE WORLD
lies in the quality of the cars actually delivered to the consumer.
American buyers report that Toyota's vehicles have among the
lowest number of defects of any in the world, comparable to the
very best of the German luxury car producers, who devote many
hours of assembly-plant effort to rectification.
LEAN PRODUCTION: TNE SUPPLY GllAlN
Assembling the major components into a complete vehicle, the
task of the final assembly plant, accounts for only 15 percent or
so of the total rnanitfacturing process. The bulk of the process
involves engineering and fabricating more than 10,000 discrete
parts and assembling these into perhaps 100 major componentsengines, transmissions, steering gears, suspensions, and so forth.
Coordinating this process so that everything comes together
at the right time with high quality and low cost has been a
continuing challenge to the final assembler firms in the auto
industry. Under mass production, as we noted earlier, the initial
intcntion was tu integrate the entire production system into one
huge, bureaucratic command structure with orders coming down
from the top. However, even Alfred Sloan's managerial innovations were unequal to this task.
The world's mass-production assemblers ended up adopting
widely varying degrees of formal integration, ranging from about
25 percent in-house production at small specialist firms, such as
Porsche and Saab, to about 70 percent at General Motors. Ford,
the early leader in vertical integration, which actually did approach 100 percent at the Rouge, deintegrated after World War 11
to about 50 percent.
However, the make-or-buy decisions that occasioned so much
debate in mass-production firms stmck Ohno and others at Toyota
as largely irrelevant. as they began to consider obtaining components Eor cars and trucks. The real question was how the assembler and the suppliers could work smoothly together to reduce
costs and improve quality, whatever Formal, legal relationship
they might have.
And here the mass-production approach-whether to make
or buy-seemed broadly unsatisfactory. At Ford and GM, the
central engineering staffs designed most of the 10,000-plus parts
in a vehicle and the component systems they comprised. The
firms then gave the dra~vingsto their suppliers, whether formally
art of the assembler firm or independent businesses, and asked
em for bids on a given number of parts of givcn quality jusually
pressed as a maximum number of defective parts per 1,000)
livered at a given time. Among al1 the outside hrms and interna1
visions that were asked to bid, the low bidder got the bu sin es^.'^
For certain categories of parts, typically those shared by
ny vehicles (tires, batteries, alternators) or involving some
cialized technology that the assembler firm didn't have (engine
mputers, for example), independent supplier firms competed to
upply the parts, usually by modifying existing standard designs
o meet the specifications o l a particular vetiicle. Again, success
pended upon price, quality, and delivery reliability, and the car
akers often switched business between firms on relatively short
In both cases, corporate managers and small-business owners
alike understood that it was every firm for itself when sales
declined in the cyclical auto industry. Everyone thought of their
business relationships as charac~eristicallyshort-term.
As the growing Toyota firm considered this approach to components supply, Ohno and others saw many prohlems. Supplier
organizations, working to blueprint, had little opportunity or
incentive to suggest improvements in the production design based
on their own manufacturing experience. Like employees in the
mass-production assembly plant, they were told in effect to keep
their heads down and continue working. Alternatively, suppliers
offering standardized designs of their own, modified to specific
vehicles, had no practica1 way of optimizing these parts, because
they were given practically no information about the rest of the
vehicle. Assemblers treated this information as proprietary.
And there were other difficulties. Organizing suppliers in
vertical chains and playing them against each other in search of
the lowest short-term cost blocked the flow of information horizontally betureen suppliers, particularly on advances in manufacturing techniques. The assembler might ensure that suppliers had
low profit margins, but not that they steadily decreased the cost
of production through improved organization and process innoThe same was true of quality. Because the assembler really
knew very little about its suppliers' manufacturing techniqueswhether the supplier in question was inside the assembler firm or
independent-it was hard to improve quality except by eslablish-
60
THEMACHlNE THATCHANGED Ti+€ WORLD
ing a maximum acceptable leve1 of defects. As long as most firms
in the industry produced to about the same leve1 of quality, it was
difficuli to raise that level.
Finally, therc was the problem of coordi~iatingthe flow of
parts within the supply system on a day-to-day basis. The inflexibility of tools in supplier plants (analogous to the iriflexibility of
the stamping presscs in the assembler plarits) and the erratic
iiatiire ol orders from assemblers responding to shifting market
demand caused suppliers to build la]-ge volumes ol one type of
part before changing over machinery to the next and to maintain
large stocks of finished paris in a warehouse so that the asseinbler
would never havc cause to complain (or worse, to cancel a contract) because of a delay in delivery. The result was high inventory
costs aiid the routine production of thousands of parts that were
later found to be defective when installed at the assembly plant.
To counteract these problems and to respond to a surge in
demand in the 1950s, Toyota began to establish a new, leanproduction approach to components supply. The first step was to
organize suppliers into functional tiers, whatever the legal, formal
relation of the supplier to the assernbler. Different responsibilities
were assigned to firms in each tier. First-tier suppliers were
responsible for working as an integral part of the product-development team in developing a new product. Toyota told them to
develop, for example, a steering, braking, or electrical system that
woiild work in harmony with the other systems.
First, they were given a performance specification. For example, they were told to design a set of brakes that could stop a
2,200-pound car irom 60 miles per hour in 200 feet ten times in
succession without fading. The brakes should fit into a space 6" x
8" x 10 " at the end of each axle and be delivered to the assembly
plant for $40 a set. The suppliers were then told to deliver a
prototype Sor testing. If the prototype worked, they got a production order. Toyota did not specify what the brakes were made of
or how they were to work. These were engineering decisions for
the supplier to make.
Toyota encouraged its first-tier suppliers to talk among themselves about ways to improve the design process. Because each
supplier, for the most part, specialized in one type of component
and did not compete in that respect with other suppliers in the
group, sharing this information was comfortable and mutually
beneficial.
Then, each first-tier supplier formed a second tier of suppliers
nder itsell. Companies in the sccond tier were assigned the job
f fabricating indiiridual pai-ts. Thesc suppliers \vere manuí'acturng specialists, usually \vithout much expertise in product engieering but with strong backgrvunds in process enginceriiig arid
lant operations.
For cxarnple, a lirst-tier supplier migllt be responsible for
manufacturing alternators. Each alteriiator has around 100 parts,
and the first-tier supplier iilould obtain al1 of these parts Srom
econd-tier suppliers.
Because second-tier suppliers werc al1 spccialists in manulacturing processes and not competitors in a specific type ol component, it was easy to group them into supplier associations so that
they, too, could exchange information on advances in rnanufactusing techniques.
Toyota did not wish to vertically integrate its suppliers into a
single, large bureaucracy. Neitlier- did it wish to deintegrate them
into completely independent companies with only a marketplace
relationship. Instead, Toyota spun its in-house supply operations
off into quasi-independent first-tier supplier companies in which
Toyota retained a fi-action of the equity and developed similar
relatiorrships with other suppliers who had been completely indeoendent. As the process Droceeded, Toyota's first-tier suppliers
acquired much of thc rest of thc equity in each other.
Toyota, for instante, today holds 22 percent of Nippondenso,
which makes electrical components and engine cornputers; 14
percent ol Toyoda Gosei, which makes seats and wiring systenis;
12 percent of Aishin Seiki, which makes metal engine parts; and
19 percent of Koito, which makes trim items, upholstery, arid
plastics. These firms, in turn, have substantial cross-holdings in
each other. l n addition, Toyota often acts as banker für its supplier
group, providing loans to finance the process machinery required
for a new product.
Finally, Toyota shared personnel wiih its supplier-group firms
in two ways. It would lend them personnel to cieal with workload
surges, and it would transler senior managers not iil linc Sor top
positions a t Toyota to senior positions in supplier firms.
'onsequently, the Toyota suppliers were independent cümpanies, with completely separate books. They were real profit centers, rather than the sham profit centers of inany vertically integrated mass-production firms. Moreover, Toyota encouraged them
to perform considerable work Sor other assemblers and for firms
in other industries, because outside business almost ali\rays gcn-
64
THE MACHINE THAT CHANGED THE WORLD
lean engineering was a drarriatic leap in productivity, product
quality, and responsiveness to changing consumer demand.
LERH PRDOUGTlDN AND GHRNGING GDNSUMER OEMRND
The new Toyota production system was especially well suited to
capitalize upori the changing demands that consumers \vere placing on their cars and upon changing vehicle technology. By the
1960s, cars and light trucks were increasingly a part of daily lile
in deireloped countries. Almost eveqone, even people with no inIierent interest ir1 cars, depended on them to get through the day.
Simultaneously, vehicles were acquiring features that made
them quite impossible for ihe average user to repair. The putty
knife and wrench that could lix almost anything that could go
wrong with a Mode! T were OS little use by ?he 1980s for a broken
engine-management computer or antiskid braking system.
Also, as households began to acquire more than one vehicle,
people no longer wantcd just tlie standard-size car or truck. The
market began to Sragment into many product segments.
For the Toyota production system, these developments were
al1 blessings: Consumers began to report that the most important
Eeature OS thcir car or tmck was reliability. It had to start every
morning and could never leave the user stranded. Vehicle malfunctions were no longel- a challenge for the home tinkerer, but were
inexplicable nightmares, even for owners with considerable mechanical skills. Because the Toyota system could deliver superior
reliability, soon Toyota found that it no longer had to match
exactly the price of competing mass-production products.
Furthermore, Toyota's flexible production system and its ability to reduce production-engineering costs let the company supply
the product variety that buyers wanted with little cost penalty. In
1990, Toyota is offering consumers around the world as many
products as General Motors-even tliough Toyota is still Iialf GM's
size. To change production and model specifications in massproduction firms takes many years and costs a Sortune. By contrast, a preeminent lean producer, such as Toyota, needs half the
time and effort required by a mass-producer such as GM to design
a new car. So Toyota can ofter twice as many vehicles with the
same development budget.
THERISE OFLEAN PRODUCTION
65
Ironically, most Western conipanies concluded that the Japanese succeeded because they produced staiidardized products in
ultra-high volume. As recently as 1987 a manufacturing manager
in Detroit confided in an interview with members of our project
that he had found the secret of Japanese success: "They are
making identical tin cans; if 1 did that 1 could have high quality
and low cost, toa." This illusion stems from the facr that the
Japanese companie initiallv minimized distribution costs bv
focusing on one or two produci categories in each export market.
Hotvever, the total nroduct portfoiio of tlie Japanese firms has
always been broader. and they have stcadily increased their product range in every world market. Today they offer almost as many
models as al1 of the Western firms combined, as we will see in
Chapter 5.i7 In addition, their product varietv continues to grow
rapidly even as Western firms stand still on average and actualiy
reduce the number of different models made in each of their
Sactories. Ford and GM, for example, have been "focusing" their
assembly plants toward tlie goal of one basic produc~in each
plant. By contrast, the Japanese transplants in North America al1
build two or three products.
Because product lives no\\/ average just four years, the average
production volume of a Japanese car ovcr the period of its production is now one-quar~erthat of the Western mass-market producers, and ¡he gap is widening. That is, the Japanese currently
make, on average, 125,000 copies of each of their car models each
year. The seven Western high-volume companies make nearly
twice as many. However, the Japanese keep models in production
four years on average, while the Western companies keep them in
production almost ten years. This means that during the life oE a
model the Japanese make 500,000 copies (125,000 times 4), while
the Western companies make 2 million (200,000 times lo), a fourto-one diflerence.
Even more striking, Japanese producers such as Toyota are
already producing a t only two-thirds of tlle life-of-the-model production volume of European specialist firms, such as Mercedes
and BMW. Indeed, with the arrival of a host of new Japanese niche
cars, such as ?he Honda NS-X, the Japanese may be able to do
what mass-production firms never could: attack the surviving
craft-based niche producers, such as Aston Martin and Ferrari, to
bring the whole world into the age of lean production.
66
1
THE MACHINE THAT CHANGED THE WORLD
1
LEAN PRODUCTION: DEALMG MTH THE CUSMH
1
Al1 of the variety available liom lean production would be for
naught iE the lean producer could not build what tlie customer
wanted. Thus íroin an early date Eiji To.yoda and his marketing
expert, Shotaro Kamiya, began to think about the link between
the production system and the consumer.
For Henry Ford this link had been very simple: Because there
was no product variety and because most repairs could be handled by the owsler, t l ~ job
e of the dealer was simply to have enough
cars and spare parts in stock to supply expected demand. In
addition, because demand in the American car market fluctuated
wildly from the earliest days of the industry, the assembler tended
to use the dealer as a shock absorber to cushion the factory from
the need to increase and reduce production continually. The
result, fully in place by the 1920s, was a system of small, iinancially independent dealers who maintained a vast inventory of
cars and trucks waiting lor buyers.
Kelations between the factory and the dealer were distant
and usually strained as the factory tried to lorce cars on dealers
to smooth production. Relations between the dealer and the
customer were eq~iallystrained because dealers continually adjusted prices-made deals-to adjust demand with supply while
maximizing profits. As anyone who has bought a car in North
America or Europe knows, this has been a system marlced by a
lack of long-term commitment o11 either side, which maximizes
leelings of mistrust. In order to maximize bargaining position,
everyone holds back is~formation-the dealer about the product,
the consumer about his or her triie desires-and everyone loses in
the long term.
Kamiya had learned this system by working in General Motor's Japanese distribution system in the 1930s, but it seemed
broadly unsatisfactory. Therefore, after the war he and Toyoda
began to think of new ways to distribute c a r ~ . "Their solution,
worked out gradually over time, was to build a sales network very
similar to the Toyota supplier group, a system that had a very
diflerent relation with the customer.
Specifically, the Toyota Motor Sales Company'Quilt up a
network of distributors, some wholly uwned and some in which
Toyota held a small equity stake, who had a "shared destiny"
67
ith Toyota. These dealers developed a new set os techniques that
yota came to cal1 "aggressive selling." The basic idea was to
velo^ a long-term. indeed a life-lone,
- relation between the
ler, and the buyer by building the dealer into
tem and the buyer into the product developThe dealer becaine part of the production system as Toyota
dually stopped building cars in advance Sor unknown buyers
build-lo-order system in which the dealer was
first step in the kalzhalz system, sending orders for presold cars
the Sactory for delivery to specific customers in t ~ v o10 three
eks. To make this workable, however, the dealer had to work
he factory to sequellce orders iil a way the factory
odate. While Ohno's production system was ret a t building products to specihc order, it could
rge surges or troughs in total demand or abnipt
etween products that could not be built with
for example, between the largest and smallest
cars in the product range or between cars and trucks.
Sequencing orders was possible in turn because Toyota's sales
staff did not wait in the showroom for orders. Instead they wenl
directly to customers by making house calls. When demand began
to droop they worked more hours, and when demand shifted they
concentrated on households they knew were likely to want the
type of car thc Sactory could build.
The latter was possible because of a second feature ol aggressive selling-a massive dara base on households and their buying
preferentes that Toyota gradually built up on every household
ever showing interest in a Toyota product. With this information
in hand, Toyota sales staff could target their eFSorts to the most
lso could incorporate the buyer into the ~ r o d u c t
ss and in a very direct way. Toyota focused
relentlessly on repeat buyers-critica1 in a country where government vehicle inspections, the famous slzolten, resulted in practically every car being scrapped after six years. Toyota was determined never to lose a fornier buyer and could minimize he
chance of this happening by using the data in its consumer data
base to predict what Toyota buyers would want next as their
incomes, Samily size, driving patterns, and tastes changed. Unlike
mass-producers who conduct product evaluation "clinics" and
other survey research on randomly selecied buyers-buyers pre-
68
THE MACHINE THAT CHANGED THE WORLD
sumed to have little "brand 1oyaIty"-Toyota went directly to its
existing custoiners in planning new producis. Established customers were treated as menibers of the "Toyotri family," and
brand loyaliy became a salieiit leature oí' Toyota's lean production
systcrii.
Toyota had fully worked out tbe principies oF lean production by
the early 1960s. The other Japanese auto firrns adopted most of
them as ivell, although this took many years. For example, Mazda
did not fully embrace Ohno's ideas for running factories and the
supplier system until it encountcred a crisis in 1973, when export
deniand for its fuel-hungry Wankel-engined cars collapsed. The
first step of the Sumitomo group in ofíering help to Mazda \vas to
insist that the company's Hiroshinia production complex rapidly
remake itself in the image of Toyota City at Nagoya.
What's more, not al1 firms became equally adept at operating
the system. (One our most important objectives in this volume is
to educatc the public to the fact that some Japanese firms are
leaner than others and that several of the old-íashioned massproduction firms in the West are rapidly becoining lean as well.)
Nevertheless, by thc 1960s the Japanese firms on average had
gained aii eilormous advantage oler rnass-producers elsewhere
and were able for a period of twenty years to boost their share
of world motor vehiclc productioii steadily by exporting from
their highly focused production complexes in Japan, as shown in
Figure 3. l .
This path of export-led devclopment came to an abrupt halt
after 1979, as the world economy slumped, trade imbalances with
North America and Europe reached iinmanageablc proportions,
and trade barriers \vere erected. In the 1980s the world was at the
same point in the diffusion of lean production that it \vas with
mass production in the 1920s: The lcading practitioners of the
new inethod are now of necessity attempting to increase their
share of the world markct through direct investment in North
America and Europe (as shown in the checked area in Figure 3.1)
rather than through ever growing exports of finished units. Meanwhile, American, Europcari, aiid cven Korean firms-oftcn accom-
69
JHE RlSE OF LOlN PRODUCJION
Japanese Share o1 World Motor Vehicle Production, 1955-1989
l
32
Year
Note:
lnciudes boih domestic and oii-shore produciion.
Source: Auiomoiive News Markel Data Book
plished masters of now obsolete mass production-are t-ing io
match or exceed the performance of their lean challengers.
This process is enormously exciting. It also produces cnormous tensions. There will be real losers (includiiig somc «f tbe
smaller and less accomplished Japanese firms) as \\-el1as winncrs,
and the public everywhere tends al1 too 1-eadily to interpret tlie
contest in simple nationalistic terms-"us" versus "tílem," "our"
country versus "theirs.
We will return to the problem of diffusing lean production in
the final chapters oE tliis book because we believe it is one of the
most important issues facing the world econoiily in the 1990s.
However, we first need to gain a deeper understanding OS rhe
elements of lean production
,S
tal image of auto production-the assembly plant where al1 the
parts come together to create the finished car or truck. While this
final manufacturing step is importan^, it represents only about 15
percent of the human effort involved in making a car. To properly
understand lean production, we must look a t every step in the
process, beginning with product dcsign and engineering, then
going lar beyond the factory to the customer who relies on the
automobile for daily living. In addition, it is critica1 to understand
the mechanism of coordination necessary to bring al1 these steps
into harmony and on a global scale, a mechanism we cal! the lean
erzterprise.
In the chapters just ahead, we proceed through each of the
steps of lean production. We begin with the part of the system
everyone thinks they understand-the factory as represented by
the assembly plant-to show systematically how very different
lean production is lrom the ideas o€ Henry Ford. We proceed to
product development and engineering, ihen into the supply system, where the bulk of manufacturing occurs. Next we look at the
system of selling cars and trucks-the end of the production
process in the world of mass production, but the beginning o€ tlie
process in lean production. Finally, \ve examine the type of global
lean enterprise needed to make the whole system work, the one
aspect ol lean production that is still not fully developed.
4
The automotive assembly plant dominales its
landscape, wherever in the world it's located.
From a distance, it is a vast windowless mass
surrounded by acres of storage areas and railway
yards. The complex shape of the building and the lack of a facade
often make it hard to know just where to enter. Once inside, the
scene is initially bewildering.
Thousands of workers in one vast building tend to streams of
vehicles moving across the floor, while a complcx network of
conveyors and belts in the lofty ceiling carries parts to and fro.
The scene is dense, hectic, noisy. On first exposure, it's like finding
oneself inside a Swiss watch-fascinating but inc~rnp~ehensible
anda little frightening, as well.
In 1986, at the outset of the IMVP, we set out to contrast lean
production with mass production by carefully surveying as many
of the world's motor-vehicle assembly plants as possible. In the
end, we visited and systematically gathered information on more
than ninety plants in seventeen countries, or about half of the
assembly capacity of the entire world. Ours .rvould prove one of
This chaptcr is based on the IMVP World Assembly Plant Surrey. The surve).
was iniiiated by John Krafcik, who was later joined by John Paul MacDuliie.
Haruo Shimada also assistcd.
THE MACHINE THAT CHANGED THE WORLD
the most comprehensive international surveys ever undertaken in
the automobile or any other industry.
Why did \ve choose the assembly plant for study? Why not
the engine plant, say, or the brake plant 01- the alternator factory?
And why so many plants in so many countries? Surely, the best
lean-production plant in Japar~and the xvorst mass-production
plant in North America or Europe would have sufficiently demonstrated the differei~cesbetween lean and mass production.
Tliree í'actors convinced us that the assembly plant was the
most useful activity in the motor-vchicle production system to
study.
First, a large part or the work in the auto industry involves
assembly. This is so simply because o i the large number of parts
in a car. Much of this assembly occurs in cornpoiierits plarits. For
example, an alternator plant will gathcr from suppliers or fabricate the 100 or so discrete parts in an alternator, tlien assernble
them ir~toa complete unit. However, it's hard to understand
assembly in such a plarit, because the final activity usually makes
up orlly a sriiall part of tlie total. In the final assembly plant, by
contrast, the sole activity is assembly-welding and screwing
several thousand simple parts and complex components iiito a
finished vehicle.
Second, assembly plants al1 over tlie world do almost exactly
the same tliings, because practically al1 of today's cars and light
trucks are built with very similar fabrication techniques. In almost every assembly plant, about 300 stamped steel panels are
spot-welded into a complete body. Then the hody is dipped and
sprayed to protect it lrom corrosion. Next, it is painted. Finally,
thousands of mechanical parts, electrical items, and bits ofupholstery are installed inside the painted hody to produce the complete automobile. Because these tasks are so uniform, we can
meaningfully compare a plant in Japan with one in Canada,
another in West Germany, and still another in China, even though
they are making cars that look very different as they emerge from
the factory.
Finally, we chose the assembly plant for study, because Japanese ellorts to spread lean production by building plants in North
America and Europe initially involved assembly plants. When we
began our survey in 1986, three Japanese-managed assembly
plants \vere already in operation iil the United States and one was
ready to open in England.
By contrast, Japanesc plants Sor engines, brakes, alternators,
UNNINGTHE FACTORY
d other components, thoiigh publicly announced Lor North
erica and Europe, were still in ihe planiiing stage. We lcnew
om experience that it's pointless to examine a company's blueints for a new plant or to look at a piant just as it starts
oduction. To see thc full difl'erence between lean and mass
oduction at the plant level, we had to compare plants operating
What about the second question \ve are often asked: "Why
tudy so many plants in so many couniries?" The answer is
imple. Lean production is now spreading from Japan tu praciially every nation. Directly in its path are the giant mass-producn plants o i the previous iridustrial era.
In every country and every company-including, we mighi
add, in the less accomplished companies in .Tapan-\ve havc fourid
an intense, even desperate, desire to know the arlswer to two
simple questions: "Where do we stand?" and "What rnust we do
to match the new competitive level required by lean production?"
/
ELASSIC MASS PRODUCTION: CM FRAllN6HLM
We began our survey in 1986 at General Motors' Frarninghanl,
Massachusetts, assembly plant, just a Few miles south olour horne
base in Boston. We chose Framingham, riot hecause i i was riearby,
but because we strongly suspected it embodied al1 tlie elements
of classic mass production.
Our first interview with ihe plant's senior managers \vas not
promising. They liad just returned Srom a tour of the Toyota-GM
joint-venture plant (NUMMI) where John Krafcik, our assembly
plant survey lcader, formerly worked. One reported that secret
repair areas and secret inveniories had to exisi hehind thc
NUMMI plant, hecause he hadn't seen enough of either Sor a
"real" plant. Another manager wondered what al1 the fuss \\las
about. "They build cars just lilce we do." A third warricd that "al1
that NUMMI talk [about lean production] is noi welcome around
here."
Despite ihis cold beginning, we found thc plaiit managemcnt
enormously helplul. Al1 over the world, as \\fe have since discovered again and again, managers and \vorkers badly warit tu Iearii
78
THE MACHINE THATCHANGED THE WORLD
where they stand and how to improve. Their fear of just how bad
things might be is in fact what often creates initial hostility.
On the plant floor, we found about what we had expected: a
classic mass-production environment with its many dysfunctions.
We began by looking down the aisles next to the assembly line.
They were crammed witli what we term indirect woi-kers-~vorkers on their way to relieve a fellow employee, machine repairers
en route to troubleshoot a problem, housekeepers, inventory runners. None of tbese workers actually add value, and companies
can find other ways to get their jobs done.
Next, we looked at the line itself. Next to each wol-k station
were piles-in some cases weeks' worth-of inventory. Littered
about were discarded boxes and other temporary wrapping material. On the line itself the work was unevenly distributed, with
some workers running niadly to keep up and others finding time
to smoke or even read a newspaper. In addition, at a number of
points the workers seemed to be struggling to attach poorly fitting
parts to the Oldsmobile Ciera models they were building. The
parts that wouldn't fit a t al1 ivere unccremoniously chucked in
trash cans.
At the end of' the line we fourid what is perhaps the best
evidence of old-fashioned mass production: an enormous work
area Full of finished cars riddled with defects. Al1 these cars
needed further repair before shipment, a task that can prove
enormously time-consuming and often fails to fix fully the problems now buried under layers of parts and upholstery.
On our way back through the plant to discuss our iindings
with the senior managers, we found two final signs of mass
production: large buffers of fiilished bodies awaiting their trip
through the paint booth and from the paint booth to the final
assembly line, ancl massive stores of parts, many still in the
railwaycars in which they had been shipped from General Motors'
components plants in the Detroit area.
Finally, a word on the work force. Dispirited is the only label
that would fit. Framingham workers had been laid off a halfdozen times since the beginning of tlie American industry's crisis
in 1979, and they seemed to have little hope that the plant could
long hold out against the lean-production facilities locating in tbe
American Midwest.
RUNNING THE FACTORY
79
CLASSIC LEAN PRODUCTION: TOYOI~IJAKADKA
ur next stop was the Toyota assembly plant at Takaoka in Toyota
ity. Like Framingham (built iu 1948), this is a middle-aged
cility (from 1966). It had a much larger number of welding and
inting I-obots in 1986 but \vas hardly a higb-tech facility of thc
t General Motors was then building ibr its new GM-10 models,
in which computer-guided carriers replaced the final assembly
The differences betweeii Takaoka and Framingham are striking to anyone who understands the logic of lean produciion. For a
start, hardly anyone was in the aisles. The armies of indirect
workers so visible at GM were missing, and practically every
worker in sight was actually adding value to the car. This fact ivas
even more apparent because Takaoka's aisles are so narrow.
Toyota's philosophy about ihe amount of plant space necded
for a given production volume is just the opposite of GM's a l
Framingham: Toyota believes in having as little space as possible
so that face-to-face communication among workers is easier, and
there is no room to store inventories. GM, by contrast, has believed that extra space is necessary to work on \gehicles needing
repairs and to store the large inventories needed io ensure smooth
production.
The final assemblv line revealed further differences. Lcss than
an hour's worth o€inventory was next to each worker at Takaoka.
The parts went on more smoothly and the work tasks were better
balanced, so that every worker worked a t about the same pace.
When a worker found a defective part, he-there are no lvomen
working in Toyota plants in Japau-xarefully tagged it and sent it
to the quality-eontrol area in order to obtain a replacement part.
Once in quality control, employees subjected the part to what
Toyota calls "the five why's" in which, as we explained in Chapter
2, the reason lor the delect is traced back to its ultimate cause so
that il will not reciir.
As we noted, each worker along the line can pul1 a cord just
above the work station to stop the line il any problem is found; a t
GM only senior managers can stop the iine for any reasoil other
than safety-but it stups frequently diie to problems with machinery or materials delivery. At Takaoka, every worker can stop the
line but the line is alniost nevcr stopped, because problems are
-
~~
THE MACHINE THAT CHANGED THE WORLD
solved in advance and the same problem never occurs twice.
Clearly, paying relentless attention to preventing defects has removed most of the rcasons for the line to stop.
At the end of the line, the dilference between lean and mass
production was even more striking. At Takaoka, we observed
almost no rework arca a t all. Almost every car was driven directly
from the line to the boat 01- the trucks taking cars to the buyer.
On the way back through ¡he plant, we observed yet other
differences between this plant and Framingham. There were practically no buffers beiween the welding shop and paint booth and
between paint and final assembly. And there were no parts warehouses a l all. lnstead parts were delivered directly to the line at
hourly intervals from the supplier plants where they had just been
made. (Indeed, our initial plant survey form asked how many days
of inventory were in the plant. A Toyota manager politely asked
whether there was an error in translation. Surely we meant
ininutes of inventory.)
A final and striking difference with Framingham was the
morale of the work force. The work pace was clearly harder at
Takaoka, and yet there was a sense of purposefulness, not simply
of workers going through the motions with their minds elsewhere
under the watchful eye of the foreman. No doubt this was in
considerable part due to the fact that al1 of the Takaoka workers
were liletiine employees of Toyota, with fully secure jobs in return
for a Eull commitment to their work.'
U BOX SCORE: MASS PRODUGTlDN VERSUS LEAN
When the team had surveyed boih plants, we began to construct
a simple box score to te11 us how productive and accurate each
plant \vas ("accurate" here ineans the number of assembly defects
in cars as suhsequently reported by buyers).' It was easy to
calculate a gross productivity comparison, dividing the number
of hours worked by al1 plant employees by the number of vehicles
produced, as shown in the top line of Figure 4.1' However, we had
to make sure that each plant was performing exactly the same
tasks. Otherwise, we wouldn't be comparing apples with apples.
So we devised a list of standard activities Sor both plantswelding of al1 body panels, application of three coats of paint,
installation of al1 parts, final inspection, and rework-and noted
RUNNING THE FACTORY
CM Frzmingham
40.7
18.0
31
16
130
-15
8.1
2 weeks
ore:
ioyoia Bkanha
4.8
2 hours
Gross assembly hours per car are calculated by dividing total hours oi elfort in tlie plant by
fhe total number of cars produced.
"Adjusted acsembly hours per car" incorporates the adjustments in standard activities and
product attributes described in the text.
Defects per car were estima?ed from the J. D. Power lnitial Quaiity Survey ior 1987.
Assembiy space per car is square feet per vehicie per year, corrected ior vehicle size.
ramingham did only half its own welding and obtained many
ewelded assemblies from outside contractors. We niade an adstment to reflect this fact.
We also knew it would make little sense to compare plants
ssembling vehicles of grossly differeilt sizes and with differing
mounts of optional equipment, so we adjusted the amount of
ffort in each plant as if a standard vehicle of a specificd size and
ption content were being assembled."
When our task was completed, an extraordinary iii?ding
merged, a s sborvn in Figure 4.1. Takaoka was almost twice as
oductive and three times as accurate as Framingham in perrming the same set of standard activities on our standard car.
82
THE MACHlNE THAT CHANGED THE WORLD
tion, as practiced by GM, ¡o classic lean production, as performed
by Toyota, really deserves the terin yevolution. After all, Ford
managed to reduce direct assembly effort by a factor of nine at
Highland Park.
In í'act, Takaoka is in some ways an even more impressive
achievement than Ford's at Highland Park, because it represents
an advance on so many dimensions. Not only is effort cut in half
and delects reduced by a factor of three, Takaoka also slashes
iilventories and manufaciuring space. (That is, it is both capital
and labor-saving compared with Framingham-style mass production.) What's more, Takaoka is abie to change over in a few days
from one type of vchicle to the next generatiun of product cvhile
Highland Park, with its vast array ol dedicated tools, was closed
for months in 1927 when Ford sxvitched from the Mode] T to the
new Model A. Mass-production plants continue to close Sor
months while switching to new products.
IKFUSlNk LfRN PRUOUCTION
Revolutions in manufacture are useful only if they are available
to everyone. We were therefore vitally interested in Iearning if the
new transplant facilities being opened in North America and
Europe could actually institute lean production in a different
environment.
We knew one of the Japanese rransplants in North America
ver" well. of course. because of IMVP research alfiliate John
~ r a f c i k ' stenure there. The New United Motor Mannfacturing Incl
(NUMMI) plant in Fremont, California, is a joint venture between
the classic mass-producer, GM, and the classic lean producer,
Toyota.
NUMMI uses an old General Motors plant built in the 1960s
to assemble GM cars and pickup trucks for the U.S. West Coast.
As GM's market share along the Pacific Coast slipped steadily, the
plant had less and less work. It finally closed its doors for good in
1982. By 1984 GM had decided that it needed to learn about lean
production from the master. So it convinced Toyota to provide
the management for a reopened plant, which k o u l d produce
small Toyota-designed passenger cars for the U.S. market.
NUMMI was to make no compromises on lean production.
The senior managers were al1 from Toyota and quickly imple-
ented an exact copy of thc Toyota Production System. A key
tion toward this end was the construction of a new stamping
nt adjacent to the body-welding area, so tliat body pancls
uld be siamped in small lots just as they xvei-e needed. By
ntrast, the old Fremont plant had depended on pariels supplied
rail from GM's centralized stamping plants i n ihr Midwesi.
ere they werc stamped out by the million ori dedicatcd presses.
The Unitcd Autornohile Workers Union also coopcraied to
ake lean production possible. Eighty percent of the NUMMI
ork Sorcc consistcd of ~vorkersSormerly employed by GM at
emont. However, iii place of ¡he usual union contraci wiih
ousands of pages of íine print defining narro\\' j«b categorics
d other job-control issues, the NUMMI contract providcd h r
nly two categories ol xvorkers-assemblcrs and iechnicians. The
nion agreed as well that al1 its meinbers should work i i l smali
teams to get the job done with the least effort and higliest quality.
By the fall of 1986, NUMMl was rurlning full blast. And we
were ready to compare it with Takaolca and Framingham, as
hown in Figure 4.2.
We Sound that NUMMI matcbed Takaoka's qualiiy and nearly
matched its produciivity. Space utilization was no¡ as efíicient
because OS tlie old GM plailt's poor layout. Iiiventory was also
considerably higher than at Takaoka, because almost al1 the parts
were transported 5,000 miles across the Pacific rather than five ur
ten miles kom neighboring supplier plants in Toyota City. (Even
so, NUMMI was able to run with a txvo-day supply OS p:rrts, \\¡hile
Framingham needed two weeks' worth.)
NUMMl Fremont, 1987
CM Framhgham
Assembly Hours
per Car
Assembly Defects
per 100 Cars
Assembly Space
per Car
Toyota Bkaoka
MUMMi Fremont
31
19
135
45
7.0
2 weeks
2 hours
2 days
THE MACHlNE THATCHANGED THE WORLD
It was clear to us by the end of 1986 that Toyota had truly
achieved a revolution in manufacturing, that old mass-production
plants could not compete, and that the ncw best way-lean production-could be transplantetl successiully to ne\v environments, such as NUMMI. Given these findings, we were hardly
surprised by subsequent events: Takaoka continues to improve,
now with much additional automation. NUMMl is also getting
continually better and a sccond line is heing added to assemble
'kyota pickiip trucks. Framingham was closed forevcr in the
summer of 1989.
RUNNING THE FACTORY
ssembly Plan1 Productiviiy, Volume Producers, 1989
SURlEYlNC THE WORLD
Once we finished our initial survey, we were determined to press
ahead on a survey of the entirc world. We werc motivated partly
by the fact that the companies and governments sponsoring us
wanted to know where they stood and partly by the knowledge
that a survey ol three plants could not answer a number of
questions about \\,ha!. roles automalion, manulacturability, product variety, aiid rrianagement practices play in the success of
manufacturing.
Ho~vever,we soon realized that we would have to conceal
company and plant names when we reported our findings. Many
companies were willing to grailt us access to their plants only on
the condition that \ve would not reveal plant names in our results.
We have respected their wishes and in this book identify plants
only wlicn the company has agreed.
After lour more years of researcli, we have found the following
about productivity and quality (or accuracy) across the world, as
summarized in figures 4.3 and 4.4.
These findings are not at al1 what we had expected. We had
anticipated al1 of the Japanese firms in Japan to be roughly
coniparable in performance-that is, equally lean. Further, we
had expected al1 of thc American plants in North America and the
American- and European-owned plants in Europe to perform at
about the same level with little variation and to trail the average
Japanese plant by about the same degree that Framingham
trailed Takaolca in 1986. Finally, we expected the assembly plants
in developing countries to be markcd by low productivity and low
quality. The reality is dilferent.
Paient ~ o i a t i o n ~ ~Lacalion
~ant
Note:
Volume psoduceis include !he Ameiican "Big Three"; Fiat, PSA. Renauit, ai>d Folkswagen in
Europe; and al1 of the companies from Japan.
JIJ
J!NA
= Japanese-owned plants in Japan.
= ~apanese-ownedplants in Nortti Arnesica, including ioint Venture piants With
USiNA
USBJIE
EIE
NIC
= Ameiican-owned plants in Narth America.
= Amesican- and Japanese-owned plants in Euiope.
= European-owned piants in Europe.
= Plants in newly industrializing countries: Mexica. Brazil. Taiwan. and Karea.
American firms.
What we find instead is that there is a considerable range of
in Japan, indced a difference of two to
productivity
one between the best plant and the worst in both productivity
and quality. The differences along other dimcnsions-space utilization, level of inventaries, percentage ol the Factory devoted to
rework area-are much less, but there is still variation.
In North America, \ve quickly discovered that Framingham
was in fact the worst American-owned plant. Average Big Three
performance in late 1989 was mucl-i better-48 percent more
effort and 50 percent more defccts, c o m ~ a r e dwith the Framingham/Takaoka gap in 1986 of nearly twice the effori and three
THE MACHlNE THAT CHANGED THE WORLD
89
fact, come plants in Japan are not particularly lean, and a number
of Japanese-owned plants in North America are now demonstrating that lean production can be practiced far away from Japan. At
the same time, the best American-owned plants in North America
show that lean production can be implemented fully by Western
companies, and the best plants in the developing countries show
that lean production can be introduced anywhere in the world.
rocess. So the actual amount of effort expended in each plant is
ctually much greater than that shown in figures 4.5 and 4.6. In
dition, we adjusted for absenteeism, which nins a t 25 pcrcent
many of these European plants compared with 5 percent or
ss in Japan. The hours in our table represent hours actually
ked, not hours on the payroll.
Our findings were eye-opening. The Japanese plant requires
-half the effort of the Amcrican luxury-car planis, hall the
rt of the hest European plant, a quarter of the effort of the
rage European plant, and one-sixth the effort of the worst
pean luxury-car producer. At the same time, the Japanese
greatly exceeds tBe quality leve1 of al1 plants except one in
urope-and this European plant requires four times the e f h t
the Japanese plant to assemble a comparable product. No
88
The productivity and quality data in figures 4.3 and 4.4 are only
for mass-market cars, that is, Fords but not Lincolns, Toyotas but
not Lexus, Volkswagens but not Mercedes. From the outset, we
believed that assembly plants are al1 pretty much the same in
what rhey actually do, no matter how prestigious the brand
they're putting together. The same type of robots, indeed often
identical models from the same manulacturer, make both the
Volkswagen and the Mercedes body welds. Paint is applied in
practically identical paint booths, and final assembly involves the
installation, largely by hand, of thousands of parts as the vehicle
moves along a lengthy assembly iine. The real dilference betwveen
the mass-market and the luxury car is that the latter may have a
thicker gauge of steel in its body, extra coats of paint, thicker
insulation, and many more luxury add-on features.
While obvious to us, this idea is not universally accepted even
in the auto industry and is certainly not the view of the broader
public. Repeatedly, executives told us that our productivity and
quality findings might be correct for the average car and light
tnick, but "luxury cars are different."
We set out to find out for certain by conducting a special
world survey oE assembly plants making luxury cars. We went to
the Japanese large-car plant that we believe, based on our survey
of the same company's mass-market car plants, to be the best in
the world. In North America, we looked at the Lincoln and
Cadillac plants. In Europe we visited Audi, BMW, Mercedes, Volvo,
Rover, Saab, and Jaguar. In each case, we carehilly standardized
the tasks being undertaken and the specifications of the vehicle,
so that we were in fact askine how much effort each plant would
need to perform standard assembly steps on the smaller and less
elaborate standard car and how many errors it would make in the
-~
~
O Weighted Average
(ii
iiegion
te:
"Luxury cars" include those produced by ¡he European "specialist pr0ducers"-DaimlerBenz, BMW, Volvo. Saab, Rover, Jaguar. Audi, and Alfa Romeo-and by Cadiilac and Lincoln
in Norfh Arnerica. The Japanese luxury category includes the Honda Legend, the Toyota
Cressida, and the Mazda 929, the three most expensive sedans being built by the Japanese
cornpanies for export in 1989. The Toyota Lexus and Nissan lnfiniti models were toa recent
!o include.
ource: iMVP World Assembly Piant Survey
nd that a third of the total effort involved in assembly occurred
his area. In other words, the German plant was expending
re effort to fix the problems i t had just created than the
U Weighted Average
k properly in the end.
Certainly, these workers are highly skilled and the work they
ever, from the standpoint o€ the lean producer this is pure
a-waste. Its cause: the failure to design easy-to-assemble
13)
Region
Note:
"Luxury cars" are as defined in Figure 4.5.
Source: IMVP World Assembly Plant Suney, based on data supplied by J. D. Power and Associates.
S
wonder the Western luxury-car producers are terrified by the
arrival of Lexus, Infiniti, Acitra, and the Japanese luxury brands
still to come.
In reviewing these data, many readers may wonder if the
difference lies in greater product variety and lower productio
scale in Europe. Certainly our mental image o€ these companies
is tliat of low-volume craft production. In fact, this is simply not
true. The European plants, with one exception, produce at the
same volume as the mass-producers we looked at earlier, and in
most cases produce a less complex mix of products than the
Japanese luxury car plant we surveyed.
When we visited the high-quality but low-productivity European plant we just mentioned, we didn't have to go far to find the
basic problem: a widespread conviction among managers and
workers that they were craftsmen. A i the end of the assembly line
was an enormous rework and rectification area where armies of
technicians in white fahoratory jaclcets labored to bring the finished vehicles up to the company's fabled quality standard. We
Our advice to any company practicing "craftsmanship" of
sort in any manufacturing activity, autornotive or otherwise,
at the source. Otherwise lean competitors wilI overwhelm
HE IMVP WORUl ASSEMBLY PLAN1 SURVEY IN SUMMARY
e 4.7 summarizes a number of dimensions of current worldperformance of the volume producers at the assemhly plant
ount of training given new assemhly workers.
92
JHE MACHlNE JHATCHANGED JHE WORLD
Japanese
in Japan
Japanese
ni ((orfh amcrica
O USINA
Períormance:
Productivity (hourslveh.)
'uality (assembly defects:100 vehicles)
A
NIC
O JiNA
81 JIJ
A E
Layout:
Space (sq. :t.lvehiclelyear)
Size o: Repair Area (as % of assembly
space)
Inventaries (days for 8 sample parts)
O
Work Force:
% o: Work Force in Teams
Job Rotation (0 = none, 4 = frequent)
SuqqestionsIEm~lovee
~ u m b eof
r Job ~lacses
Training of New Production
Workers (hours)
Absenteeism
11.9
8.7
67.1
14.8
380.3
370.0
46.4
173.3
$0
(i
20
40
SO
80
rao
izo
$40
roo
iao
1
>"o
Ouality (ass'y de!ectsIlOO vehicies)
Automation:
WelUing (% of direct steps)
Painting (% of direct steps)
Assembly (% of direct steps)
One additional and very important finding of the survey bears
note: the relation between productivity and quality. When we first
began the survey and correlated productivity with quality in al1
plants, we found almost no relationship. What's more, this did
not c h a n ~ eover time. In Figure 4.8, showing the relationship
across the world at the e n d of 1989, the c&relation between
productivity and quality is .15.
This seemed puzzling. We thought it should either be negatively correlated-plants with high quality should require more
effort to achieve this, as Western factory managers had long
thought-or it should be positively correlated-quality should be
,'
free," as many writers on Japanese manufacturing had suggested. The answer to the puzzle, as a moment's examination of
CmlWC TU LEAN
have periodically reviewed our survey findings with practily al1 the world's motor-vehicle producers, the main sponsors
the IMVP. So the figures we report here don't come as a surprise
o these companies and are now generally accepted as an accurate
ummary of the general state of competition at the lactory level.
However, determining who stands where in world compe
tion differs from explaining precisely what the also-rans need t
do to catch un. As we have reviewed our data with these comp
nies, their executives and managers have questioned us on Eou
points in particular.
First, they ask whether automation is thc secret. Our answe
is that it is and it isn't. Figure 4.9 shows the relation between th
fraction of assembly steps that are automated-either by roboti
or more traditional "hard" automation-and the productivity
plants. There is clearly a downwarci slope to the right-mor
automation means less effort. (Stated another way, higher leve
~-
L -
~
-~~
~
~.
. , . : ,
h
.
ni+?,,
,r,Tn';
,*
rr,rral.,,inn
(-
67) ,i,i
hinher
levels of effort.) We estimate that on average automatio
accounts for about one-third of the total differenc'e in productivit
.
between plants.
However, what is truly striking about Figure 4.9 is that at any
leve1 of automation the difference between the most and leas
efíicient plant is enormous. For example, the least autom
Japanese domestic plan1 in the sample (with 34 percent of al1
steps accomplished automatically), which is also the most e
cient plant in the world, needs half the human effort of o
comparably automated European plant and a third the effort
another. Looking farther to the right in Figure 4.9, we can see t
the European plant that is the most automated in the world (with
48 percent of al1 assembly steps doneby automation) requires 70
percent more effort to perform our standard set of assembly tasks
on our standard car than is needed by the most efficient pl
with only 34 percent automation.
The obvious question is, how can this be? From our surve
findings and plant tours, we've concluded that high-tech plan
that are improperly organized end up adding about as many
indirect technical and service workers as they remove unskilled
direct workers from manual assembly tasks.
What's more, they have a hard time maintaining high yield,
because breakdowns in the complex machinery reduce the fraction of the total operating time that a plant is actually producing
vehicles. From observing advanced robotics technology in many
plants, we've devised the simple axiom that lean organization
must come before high-tech process automation if a company is
to gain the full benefit.7
The also-rans' second question is, Does the manufacturability
(ease of assembly) of the product make the difference, rather than
ersus Productivity, Volurne Producers, 1989
A NIC
50
O JINA
R JIJ
-
A
A
"
A E
+
A
A~stiaiia
A
A,
110 -
A
A
+
+
h
+*
A
&
A A
A
B
-0
A
1
A
A
AA
O f f
osl.
@,O
A
.ms
B
Automatian (% of aulomaled ass'y steps)
e operation of the factory? Understandably, union leaders have
e United States, engaged us in a dialogue on this point through-
96
JHE MACHlNE JHATCHANGED JHE WORLD
Answering the manufacturability question definitively is
ficult, because we would need to perform what auto makers ca
tear-down analysis on every car being assembled in every pl
we surveyed. Only then could we see how many parts the car
and how easily they can be assembled. This analysis woul
staggeringly expensive and time-consuming. So we can repo
only some interesiing but partial evidence that ~nanufacturability
is indeed very important.
One piece of evidence is a survey we conducted in the spring
of 1990 oT the world's auto makers! We asked them to rank a11 the
other auto makers in terms of how manufacturable their products
are a t the assembly plant. They were to base their ranking on
tear-down studies that car companies conduct as part of their
competitive assessment programs. (Strange as it may seem, the
first production models of any new car are unlikely to reach
consumers. Instead, competitors buy them, then immediately tear
them apart for competitive assessment.) The results the manufacturers reported are shown in Figure 4.10.
We can't confirm the accuracy of these findings, because we
don't kiiow how much tear-down analysis companies do or how
well they do it. When we began our assembly-plant survey, we
wcre amazed to discover that very few car companies conducied
systematic benchmarking studies of their competitors. Nevertheless, the companies responding were in close agreement on which
producers design the most manufacturable designs, and the findings correlate nicely with company performance on our productivity and quality indices. This suggests that manufacturabili
conducive to high performance in the factory.
Further evidence comes from a recent comparison Genera
Motors rnade between its new assembly plant a t Fairfax, Kansas
which makes the Pontiac Grand Prix version of its GM-10 model,
and Ford's assembly plant for its Taurus and Mercury Sable
models near Atlanta. This comparison was based on tearing down
both cars, then using shop manuals to reconstruct the assembly
process.
GM found a large productivity gap between its plant and the
Ford plant-both make cars in the same size class, with similar
levels of optional equipment and selling in the same market
segment. After careful investigation, GM concluded that 41 percent of the productivity gap could be traced to the manufacturability of the two designs, as shown in Figure 4.1 1. For example,
the Ford car has many fewer parts-ten in its front bumper
:
Average RanX
nangc nf
Ranhings
2.2
3.9
4.8
5.3
5.4
5.6
6.4
6.6
8.7
10.2
11.3
12.7
13.5
13.9
13.9
14.0
16.4
16.6
18.6
1-3
1-8
3-6
2-1 1
4-7
2-8
3-9
2-1 0
5-1 1
7-1 3
9-1 3
10-15
9-1 7
12-17
10-1 7
11-1 6
13-1 8
14-18
17-19
These rankings were compiled by summing responses to a survey of the nineteen maior
assembler firms. Eight firms returned the survey in usable form-two American, four
European, one Japanese. and one Korean. The firms were asked lo rank al1 nineleen firmS
"according to how good you think each company is at designing products that are easy for
an assembly plant lo build."
9%
ZO/ó
41%
48%
100%
ource: Generai Motors
THE MACHINE THATCHANGED THE WORLD
99
compared with 100 in the GM Pontiac-and the Ford parts fit
together more easily. íThe other major cause of the productivity
gap was plant organizational practices of the type we have
discussed. The GM study found that the leve1 of automatio
which was actually much higher in the GM plant-was not
factor in explaining the productivity gap.)
Ease of manufacture is not an accident. Rather, it's one of t
most important results of a lean-design process. We'll look at th
point more carefully in Chapter 5 .
A third question that often crops up when we review ou
survey findings with companies is product variety and compl
ity. The factory manager we encouritered in Chapter 3, w
maintained he could compete with anyone if he could only foc
his factory on a single standardized product, is typical of m
Western managers. This is certainly an interesting idea and it
a simple logic to commend it.
However, in our survey we could find no correlation at a
between the number of models and body styles being run down a
production line and either productivity or product quality. We
tried a different approach by comparing what was being built in
plants around the world in terms of "under-the-skin" complexit
This was a composite measure composed of the number of ma
body wire harnesses, exterior paint colors, and engineitransmission combinations being installed on a production line, plus tbe
number o€ different parts being installed and the number of
different suppliers to an assembly plant. The results were even
less assuring to those thinking that a focused factory is the
solution to their competitive problerns: The plants in our survey
with the highest under-the-skin complexity also had the highest
productivity and quality. These of course were the Japanese
plants in J a ~ a n . ~
e specific aspects of plant operations that account for up to half
the overall performance difference among plants across the
rld? And how can these be introduced?
he truly lean plant has two key organizational features: It
nsfers the inaximum ntcmber oftasks und responsibilities to thosc
rkers actuallj~adding valtle to tlze car on tlze line, and it has in
ce a system for detecting defects that qtiickly traces every problem,
ce discovered, to its ultimate cause.
This, in turn, means teamwork among line worlcers and a
ple but comprehensive information display system that makes
ssible for everyoile in the plant to respond quickly io proband to understand the plant's overall situation. In oldned mass-production plants, managers jealously guard inrmation about conditions in the plant, thinking this knowledge
the key to their power. In a lean plant, such as Takaoka, al1
rmation-daily production targcts, cars produced so lar that
equipment breakdowns, ~ersonnelshortages, overtime reirements, and so forth-are displayed on azdoii boards (lighted
ctronic displays) that are visible from every work station. Every
e anything goes wrong anywhere in the plant, any employee
ho knows how to help runs to lend a hand.
So in the end, it is the dynamic work team that emerges as
e heart of the lean factory. Building these efficient teams is not
e. First, workers need to be taught a wide variety of skillst, al1 the jobs in their work group so that tasks can be rotated
d workers can fill in for each other. Workers then need to
quire many additional skills: simple machine repair, qualityecking, housekeeping, and materials-ordering. Then they need
couragement to think actively, indeed proactively, so they can
vise solutions before problems become serious.
Our studies of plants trying to adopt lean production reveal
at workers respond only when there exists some sense of reciproca1 oblieation, a sense that management actually valiies skilled
kers, will make sacrifices to retain them, and is willing to
egate responsibility to the team. Merely changing the organition chart to show "teams" and introducing quality circles to
d ways to improve production processes are unlikely to make
uch difference.
This simple fact was brought home to us by one of our early
dies of Ford and General Motors plants in the United States. In
Ford plants we found that the basic union-management conact had not been changed since 1938, when Ford was finally
Y8
/
1
KAN .Gl.ZAilM
A7 .1 PLAN1 EVEL
!
Those company executives, plant managers, and union leaders
accepting our conclusion that automation and manufacturability
are both important to high performance plants, but that gaining
the full potential of either requires superior plant management,
usually raise a final question that we find most interesting: What
are the truly important organizational features of a lean plant-
THE MACH/NE THATCHANGED THE WORLD
101
forced to sign a job control contract with the UAW. Workers
continued to have narrow job assignments and no formal team
structure was in place. Yet, as we walked through plant after pl
we observed that teamwork was actually alive and ~uell.Wor
were ignoring the technical details of the contract on a massive
scale in order to cooperate and get the job done.l"
By contrast, in a number of General Motors showcase plant
we Eound a new team contract in place and al1 the formal app
ratus of lean production. Yet, a few moments' observation revealed
that v e ~ ylittle teamwork was taking place and that morale on the
plant floor was very low.
How do we account for these seeming contradictions? The
answer is simple. The workers in the Ford plants had great
confidente in the operating management, who worked very hard
in the early 1980s to understand the principies of lean production.
They also strongly believed thar if al1 employees worked together
to get the job done in the best way the company could protect
iheir jobs. At the GM plants, by contrast, we found that workers
had very little confidence that management knew how to nlanage
lean production. No wonder, since GM's focus in the early 1980s
was on devising advanced technology to get rid of the workers.
The GM workers also had a fataljstic sense that many plants were
doomed ailyway. In these circumstances, it is hardly surprising
that a commitment to lean production from the top levels of the
corporation and the union had never translated into progress on
the plant floor.
We'll return to the thorny question of how lean production
can be introduced into existing mass-production factories in
Chapter 9.
ass production." They go so far as to label the lean-production
stem instituted at NUMMI in Catifornia management by stress,
se managers continually try to identify slack in the systemd work time, excess workers, excess inventories-and reove them. Critics argue that this approach makes híodenz Tir1ze.s
ok like a picnic. In Charlie Cha~lin'swidget factory at least the
orkers didn't have to think about what they were doing and try
second critique of lean production comes in the form of
t might be called "neocraftsmanship." This has been put in
ation in only a few plants in Sweden, but it draws widc
entioii across the world because it appeals to a seemingly
shakable public faith in craftsmanship.
Let's take Volvo's new UdevalIa plant in western Sweden as
example. At Udevalla, teams of Volvo workers assemhle Vol\~o's
and 760 models on stationary assembly platforms in small
k cells. Each team of ten workers is responsible for putting
ether an entire vehicle from the point it emerges from the
t oven. Looked at in one way, this system is a return full circle
enry Ford's assembly hall of 1903, which we and the rest of
world left behind in Chapter 2 . The cycle time-tlie interval
ore the worker begins to repeat his or her actions-increases
Udevella to several hours, from about one minute in a mass- or
an-production assembly plant. In addition, workers in the asmbly team can set their own pace, so long as they complete four
cars each day. They can also rotate jobs within the teams as they
desire. Automated materials-handling delivers the parts needed
for each car to the work team. Proponents o£ the Udevalla system argue that it can match the efficiency of lean-~roduction
plants while pro~~iding
a working environment that is much more
1
1
As we noted in Chapter 2, Henry Ford's sword was double-edged.
Mass production made mass consumption possible, while i t made
factory work barren. Does lean production restore the satisfaction
of work while raising living standards, or is it a sword even more
double-edged than Ford's?
Opinions are certainly divided. Two members of the United
Automobile Workers Union in the United States have recently
argued that lean production is even worse for the worker than
We strongly disagree with both points. We believe that a vital,
but often misunderstood, difference exists between tension and a
ntinuing challenge and between neocraftsmanship and Iean
oduction.
To take the first point, we agree that a ~ r o p e r l yorganized
an-production system does indeed remove al1 slack-that's why
's lean. But it also provides workers with the skills they need to
control their work environment and the continuing challenge of
making the work go more smoothly. While the mass-production
plant is often filled with mind-numbing stress, as workers struggle
to assemble unmanufacturable products and have no way to
improve their working environment, lean production
creative tension in which ~vorkershave many ways to
challenges. This creative tension in\rolved in solving
problems is precisely what has separated manual facto
hom professional "think" work in the age of mass production
To make this system work, o€course, management
its full support to the factory work force and, when
market slumps, make ihe sacrifices to ensure ,job security that h
historically been offered only to valued professionals. It truly is
system of reciprocal obligation.
What's more, we believe tliat once lean production princip
are fully instituted, companies will be able to move rapidly
the 1990s to automate most of the remaining repetitive tasks
auto assembly-and more. Thus by the end of the century w
expect that lean-assembly plants wilf be populated al
tirely by highly skilled problem solvers whose task will be to
think continually of ways to make the system run more smoothly
and productively.
The great fiaw of neocraftsmanship is that it will never reach
this goal, since it aspires to go in the other direction, back toward
an era of handcrafting as an end in itself.
\Ve are very skeptical that this form of organization can ever
be as challenging or fulfilling as lean production. Simply belting
and screwing together a large number of parts in a long cycle
rather than a smalf number in a short cycle is a very limited
vision of job enrichment. The real satisfaction presumably comes
in reworking and adjusting every little part so that it fits properly
In the properly organized lean-production system, this activity is
totaily unnecessary.
Finally, the productivity of the Udevalla system is almost
certain to be uncompetitive even with mass production, much
less lean production. We have not audited Udevalla or Kalmar,
the two Volvo plants operated on the neocraft model, but some
simple arithmetic suggests that if ten workers require 8 hours
simply to assemble four vehicles (not including welding the body,
painting it, and gathering necessary materia1s)-for
a total of 20
assembly hours per vehicle-Udevalla can hardly hope to compete
with our survey's leading lean-production plant, which requires
only 13.3 hours to weld, paint, and assemble a slightly smaller
and less elaborate vehicle.
Before leaving this point, we oSfer one final reason why lean
production is unlikely to prove more oppressive than mass pro-
production is fragile. Mass production
rs everywhere-extra
inventol-y, extra
ce, extra workers-in order to make it function. Even when
ts dontt arrive on time or many workers cal1 in sick or other
orkers fail to detect a problem before the product is massduced, the system still runs.
However, to make a lean system with no ~lack-no safety
hard.
k at all, it is essential that every worker try
oing through the motions of mass production with one's
ewhere quickly leads to disaster with
management fails to lead and the work
roca1 obligations are in force, it is quite
oductiosi will revert to mass production.
manager remarked during a plant tour:
ply lean production run by tl-ie rule book,
tiative and responsibility to continuall~
l-his last point raises come profound questions about ihe
ction across the whole xvorld, a Lwic that
n in Chapter 9. However, a t this point,
gic of lean production from the assembly
development. As we'll see, the nature of the
,,dern motorcar-a
highly complex product with more than
highly complex design and engineering
ery other aspect of production, the lean
approach to coordinating this system is fundamentally different
from that of mass production.
motion the extraordinarily complex and expensive process ol
veloping a new car.
Al1 large automobile companies-mass production or leance the same basic problem in developing a new product. A
mber of hnctional departments-marketing, power train engiering, body engineering, chassis engineering, process engineer, and factory operations-must collaborate intensixrely over an
tended period of time to develop the new cal- successfully. The
In 1981 General Motors began to plan a replacement For its just
launched front-drive A-cars and its older rear-drive G-cars, the
companv's offerings in the intermediate-size segment of the North
American market. The A-cars-the Chevmlet Celebrity, Pontiac
6000, Oldsmobile Ciera, and Buick Century-would normally have
remained in production for ten years. However, GM knew that
Ford was developing a new intermediate-size model for introduction in 1985, and the Japanese companies were thought to be
planning a much stronger presence in this segment. (Intermediate
is one of the four standard size categories traditionally used to
segment the American car market: subcompact, compact, intermediate, and full size.)
The intermediate-size segment of the market had by then
become GM's volume base, accounting for about one-third of the
company's annual car sales in North America. Senior executives
at GM concluded that they dare wait no longer than 1986 for a
new model. They knew that if they tried to run the usual ten-year
cycle with A bodies while continuing with the older G bodies,
they would fall badly behind Ford and the Japanese. So they set
This chapter is based on the research o i Takahii-o Fujirnoto, Andrew Graves,
Kentaro Nobeoka, and A~itonySheriff.
The simplest approach would be to create a totally selfntained project team colisisting ol the necessary number of
lanners and enrineers. A team manager
- could orcbestrate the
efforts of this group over a period of years until the projeci was
completed.
In fact, no company in the world, mass-producer or lean, does
this. The reasons are simple. Every company has a range of
models. mechanical components, and factories that must be
shared. Model A will share a transmission with Model B and be
built alongside Model C in the same factory. Isolating the transmission engineers and €acto- managers for Model A in a selCsufficient team would never wurk, since their efforts would soon
be at cross-purposes with teams working on models B and C.
Isolating the product planners would never work either, because
their designs might overlap with other new products in the planning pipeline. Moreover, engineers working in isolation ~vould
soon loose contact with the technical frontierof their specialty, as
it is pushed ahead by research activities in tlie functional departments. The result: Their designs would not be state-of-the-art.
In consequence, most automotive companies develop some
sort of matrix in which every employee involved in developing a
product reports both to a functional department and to a development program. The leadership challenge is managing the matrix to satisfy the needs of both the functional department and the
product-development program.
At General Motors, meeting this challenge has been particularly critical. From the 1930s to the end of the 195Os, tbe company
turned out five basic models-the Chevrolet, Pontiac, Oldsmobile,
Buick, and Cadillac. The fiie had separate chassis, bodies, and
engines but shared hundreds, or even thousands, of other partspumps, electrical components, springs, bearings, and glass. So
the development of a new model by any of the car divisions
involved a complex interaction with the other car divisions and
THE MACHINE THAT CHANGED THE WORLD
the components divisions that made the shared parts. This w
the organizational consequence of Alfred Sloan's determinatio
to share as many parts as possible to gain economies of scale.
After 1959, when GM introduced its first small cars, t
situation became still more complex. By the late 1960s the company was oflering four separate sizes of car in each car division,
except Cadillac, tvhere i t offered two models. To preserve economies oF scale while doing this, GM began to share a basic model
between its divisions, giving the model sold under each divisional
name a slightly different appearance. So the new intermediate
model introduced iil 1968 was served up as a Chevrolet Chevelle,
a Pontiac Tempest, an Olds F-85, and a Buick Skylark. These cars
had different exterior sheet metal and different dashboards and
door trims on the interiors but used exactly the same basic
components, including engines and chassis, under their metal
sbins. In other words, everything tucked out of sight was exactly
the same. To develop these products, the company now had to
coordinate the needs of four marketing divisions as well, each
wanting a different character-sporty, conservative, technologically advanced, luxurious-in their version to satis6 the expectations of traditional buyers of that division's cars.
GM's approach to developing its new model, known inside
the company as the GM-10, was standard. Senior executives
designated a program manager to take the lead in coordinating
thc functional departments involved in the development process.
Robert Dorn, chiefengineerolGM's Pontiac Division, was selected
as the GM-10 manager and given a $7-billion development budget.
He set up an office in GM's Chevrolet Division, recruited a small
peisonal staff, and went to w r k . (Since the GM system doesn't
have an OHice of Program Manager, those selected for this roie
are, in effect, nomads and must find a division to take them in.)
His first step was to get the lour car divisions to agree on the
target market of consumers for the car and the product features
thcse buyers would find most appealing. To get the job done, he
ordered large amounts of survey rcsearch among consumers and
an analysis of sales patterns in the market.
Key decisions made during this process included the physical
dimensions of the new cars, their general appearance and performance, the target markei and price (around $14,000) and
accompanying larget cost, fue1 economy (about 24 miles per
gallon), and body styles. All four divisions wanted a two-door
pe and a four-door sedan, and severa1 requested a station
Dorn's group took this information and consulted with the
Styling Center on the exact externa1 and interna1 appearance
h model. This process begins with rough sketches, proceeds
ailed rnodels in clay, then advances to actual prototypes,
'ch are shown to representative potential buyers for their
When al1 the thousands of decisions on specifications, appeare, and performance had been made, Dorn's group conveyed
details to the next group of specialists in what was then
eral Motors' Fisher Body Division and the components-engiring divisions. There engineers worked out the precise specifiions of every majar part and, more important, decided which
ts could be carried over from the existing A cars and which
Id be obtained from other GM products. Parts that could
either be carried over nor shared had to be designed from
ratch. (This advanced engineering is the most time-consuming
nd expensive part of any development program and needs to
art as soon as possible.)
By this point, Robert Dorn was becoming concerned. The GMprogram was steadily slipping its five-year timetablc, and
n's small stafl seemed powerless to make it go faster. Much of
problem stemmed from the fact that Dorn and his group were
ct coordinators, rather than managers. In vther words, they
easoned with people in order to coordinate efforts; they weren't
aders who gave orders and expected to be obeved. Wllen they
ged the engineering departments to go faster, they were met
th promises but little action. Clearly, in this matrix, each
mployee was more concerned with pleasing his or her functional
epartment boss than the GM-10 program coordinator. For exmple, if the coordinator pointed out that a feature of the engine
d changing so it would run correctly, the team represcntaorn power train engineering would stall, knowing that ihe
ngine was just fine for the cars that accounted for most of GM's
roduction volume.
As the program fe11 further behind, its problems multiplied.
e erosion of the station wagon segment by minivans caused
cellation of the station wagon version of GM-10, and thc
roduction of the Ford Taurus in 1985 caused GM to redesign
e exterior sheet metal of the GM-10 cars because senior execu-
108
THE MACHlNE THAT CHANGED THE WORLO
tives Celt they would otherwise be too similar to the Ford products.
Finally, in 1985, Dorn had enough and resigned. He was
replaced by Gary Dickenson, who faced the next major hurdle in
the GM-10 program-moving the completed product design from
Fisher Body and the components-engineering departments over
to the then General Motors Assembly Division (GMAD), which
was charged with actually manufacturing the cars. GMAD was a
monolithic organization (now disbanded in a major reorganizarion), with its own interna1 culture and career path, and Dickenson soon became as frustrated as Doi-n in trying to move the
program along within a group that was manufacturing a dozen
other major products. The timetable continued to slip.
As GM-10 was finallv readv to reach the market in 1988.
Dickenson was given a new assignment, and a third program
manager, Paul Schmidt, was given the task of overseeing the
launch. His job was to debug the four advanced-technology assembly plants designated to build GM-10 and coordinate the vast
marketing and promotional apparatus. In addition, he had to deal
with the many running chariges in the design of the cars. These
changes werr introduced after launch to increase consumer satisfaction, decrease warranty costs, and streamline factory operations.
The first GM-10 model, the Buick Regal two-door coupe,
reached buyers in the spring of 1988, seven years after the initial
decision to proceed with the project and two years after the
original deadline. The Olds Cutlass Supreme and the Pontiac
Grand Prix two-door models came next early in 1989. The Iast
model in tlie range, the Buick Regal four-door sedan, finally
reached dealer showrooms in the summer of 1990, nine years after
the GM-10 program began. Meanwhile, Ford had launched its
Taurus and Sable models as expected at the end of 1985, and
Honda had gone through two generations of its Accord model,
moving it up in size almost to match the physical dimensions of
the GM- 10 cars.
Not surprisingly, the GM-lo's, although generally agreed to
be competent, have encountered tough competition in the market.
By 1986,GM had decided that the 1.6-million-unit annual production goal for the existing A and G bodies was not realistic and had
scaled back production plans for the GM-lo's to 1 million units
annually (and from seven final assembly plants to four). The
company wanted to reach this goal by 1990. In fact, 1989 sales
were only at about 60 percent of the planned IeveI, meaning that
ven with continuing sales of some A bodies GM lost 700,000 units
f sales in its largest selling segment during the 1980s.
What's more, as we saw in the last chapter, the GM-10 cars
e neither easy nor inexpensive to make. Thus what was formerly
e of GM's most profitable areas-the intermediate segmentIonger pulls its weight. Indeed, the A bodies that GM-10 was to
lace have proved much more profitable in the late 1980s, and
company now plans to continue the production oE the Oldsobile and Buick variants indefinitely.
HE HONDR RCCDRD: LEAN PROOUCT OEVELDPWIENI
986, when the GM-10 program was already
an planning its own product for the
fourth-generation Accord, to be
a 1990 model. Since its introduction
,the Accord had been the key to Honda's success in export
and had grown steadily from subcompact to intermediincreasing incomes and farnily size of its
Honda's product development process is quite different from
i Miyoshi was appointed Large Project
for the new Accord and given powers far surpassing
dreamed of. While Honda also uses a
ich each project member is on loan from a functional
old to borrow appropriate people from
rtments and transfer them to the Accord
than coordinating, Miyoshi's task was,
, to manage. He could move the project along rapidly,
e al1 the necessary resources were under his direct control.
rd plan was finalized, it became clear that the car
rent market demands in different parts of the
, market, both a two-door coupe and a station
would be important, as would a four-door sedan. For the
se market a four-door hardtop would be needed along with
oupc, the latter to be imported from the Unitcd
pe would be served by the sedan from Japan
station wagon from the United States. In
dition, Honda needed slightly diffcrent versioiis oE each car for
separate Honda and Vigor distribution channels in Japan.
THE MACHINE THAT CHANGED THE WORLD
Honda therefore decided to subdivide its development work
into one Japanese team responsible for the basic car (including
the four-door sedan) and two sub-teariis, one in the United States
responsible for the coupe and station wagon variants, and one in
Japan responsible for the four-door hardtop. The coupe and sta
tion wagon were to be produced exclusively in the United States
at Honda's Marysville, Ohio, complex, using production tools
designed and built a: Marysville. while the sedan was to be buil
in both Japan and the United States, and the hardtop in Japa
only.
Once the product plan was set, the Honda team steame
ahead at breakneck speed and with no intenuptions. While tea
members continued to work closely with their functional depart
ments, for the reasons we just mentioned, Miyoshi and practicafly
every member of the team continued in their jobs until well after
the new model was launched on schedule in the fa11 of 1989. They
then returned to their Functional departments or were assigned to
a new product-develo~mentproject, perhaps the next generation
Accord, set for introduction in the fa11 of 1993.
While a conservative design, the Accord has been a resounding success in the marketplace, particularly in North America. In
fact, since 1989 it has been the largest-selling model in North
America, a position always held during the previous eighty years
by a GM or Ford product.
1
II S H I P S l OF PRODUCT DEYELDPMENT AROUNO THE WORLD
The GM-10 and Accord cases suggest a striking difference betweenthe lean- and mass-production approaches to product development and the consequences for competitive success. But they are
only two examples, and it would be dangerous to draw firm
conclusions on such limited, if provocative, evideuce. Fortunately,
just as we were launching our research in 1986, Professor Kim
Clark at the Harvard Business School was undertaking a worldwide survey of product-development activities in the motor industry. With the help of Takahiro Fujimoto, a Ph.D. candidate at the
Business School, Clark surveyed practically every auto assembler
in North America, J a ~ a n and
,
Western E u r o ~ e . He
' asked how
many hours of engiueei-ing effort were needed and how much time
it had taken to produce their most recent new products. Though
projects were entirely separate in inspiration, funding, and
duct, we've benefited from lengthy discussions with the Clark
am, and their work complements our global surveys of factory
actice and supply-system management.
clark and his team initially faced the same problem as we
he assembly plant: How do you ensure an apples-to-appIes
rison? Automobile develo~mentprojects can vary greatly
e size and complexity of the vehicles, the number of differeni
dy styles spun off a base model (or "platform," in car talk), the
umber of carryover parts from previous models, and the number
shared parts from other models in a producer's range. As we
ve noted, carryover and shared parts require much less engiering than totally new parts. Since thcy are aiready developed,
ey often need only minor modifications to fit into a new model.
After tbey made adjustments for these variables, the Clark
am's findiiigs were simple and striking. Based on twenty-nine
lean sheet" development projects reaching the market between
83 and 1987, Clark found that a totally new Japanese car
uired 1.7 million hours of engineering effort on average and
k forty-six months from first dcsign to customer dcliveries.'
y "clean sheet," we simply mean that these were cars with
tally new bodies, although some used canyover or shared engines.) By contrast, the average U.S. and European projects of
comparable co~nplexityand with the same fraction of carryover
and shared parts took 3 million engineering hours and coiisumed
sixtv months. This, then, is the true magnitude of the ~erformance
difference between lean and mass production: nearly a two-to-one
difference in engineering effort and a saving of one-third ir1 development time.
Perhaps the most remarkable feature oI Clark's survey is his
finding that lean product-development techniques simultaneously
reduce the effort and time involved in manufacturing. Tbis fact
turns on its head one of our most common assumptions, one that
is based on seventy years' experience with mass production:
namely, that any project can be spceded up in a crisis but that
doing so greatly increases the cost and arnount of effort.
We're al1 familiar with the refrain: "Sure 1 can get it ready
sooner, but it's going to cost you a fortune!" We suggest that
"faster is dearer" will now join "quality costs more" (an idea
debunked in Chapter 4) on the junk heap o€ ideas left over froril
the age of mass production
118
THE MACHINE THAT CHANGED THE WORL
Product Development Performance by Regional Auto Industries, Mid-1980s
Europesn
Empe
duction rate very slowly, stopping as necessaAverage Engineering Hours per New
Car (miliions)
Average Developrnent Time per New
Car (in rnonths)
Nurnber of Ernployees i n Project
Tearn
Number of Body Types per New Car
Average Ratio ot Shared Parts
Supplier Share of Engineering
Engineering Change Costs as Share
ot Totai Die Cost
Ratio of Delayed Products
Die Developrnent Time (rnonlhs)
Prototype Lead Time (months)
Time from Production Start to First
Sale imonths)
Return to Norrnai Productivity ARer
New Model (rnonths)
Return to Normal Qualitv After New
Modei (months)
io get eacli
Much more striking is the difference in quality. Japanese leanroduction plants can introduce new lean designs with only a
nts struggle for a year to get quality back to its original level,
ich is Iower than that of the Japanese to begin rvith.
1
4
5
12
Source: Kim B. Clark, Takahiro Fujimoto, and W Bruce Chew, "Praduct Development in the Worl
Auto Industry." Brookings Papers on Ecooomic Activity, No. 3, 1987: and Takahiro Fujimot
"Organizations for Effective Product Developrnent: The Case o1 the Global Motor lndustry
PhD. Thesis, Harvard Business School, 1989. Tabies 7.1, 7.4, and 7.8
When we review this box score, we can see several additional
advantages of lean design. For one, lean design results in a much
higher fraction of projects coming into production on time. Indeed, five of six Japanese projects reach the market on the timetahle laid out at the beginning of development, while only half o
American projects come in on time. The GM-10 project was wors
than average in its slippage, but haidly unusual.
Another advantage lies in the ability of the lean factor- to
absorb new products without paying a productivity penalty. Many
Western analysts have been badly misled by the slow start-up
schedules ("ramp-up rates," in car talk) oí the Japanese transplants in North America and Europe. What they fail to realize is
that these facilities are building up the social process of produc-
THE CONSEUUENCES OF LEAN DESIGN IN THE MARKETPLACE
at do companies that have mastered lean design do to take
ntage of their strength in the marketplace? Obviously, they
offer a wider variety of products and replace them more
uently than mass-production competitors. And this is exactly
at has been happening in the auto industry across the ~vorldin
In Figure 5.2 we summarize tlie number of models the Japaes were selling worldwide b e t w e n 1982 aiid
mpare this number with the total worldwide
by the U.S.-headquartered producers and the
uropean car companies (PSA, Reriault, Fiat,
er, and Volkswagen). We provide a separate calculation for
an specialist firms: BMW, Mercedes, Volvo,
The trend is striking. The Japanese firms are using their
antage in lean production to expand their product range
y renew existing products every four years.
1990 they nearly doubled their product portfofrom forty-seven to eighty-four models.
20
-
~ i a d u c tAge (yeais)
developed by General Motors, Fiat, Ford, and Volkswagen in Brazil, and the rnodels developed
by Ford and GM Holden's in Australia are not counted.
The model counl includes al1 automobiles and car-derived, front-wheel-drive minivans. It
excludes rear-wheel-drive mini-vans, sportlutility vehicles, and trucks.
A "model" is defined as a vehicle with entireiy different externa1 sheet metal from any other
praduct oHered by a company. Thus GM-10 is counted as four rnodels and Ford Taurusi
Sable is counted as two models. Two-, three-, four-, and five-door variants and station
wagon versions of the same car are coonted as one modei.
Average product age has been weiflhted by sales volume because a number of very low
volume products in Europe and Japan are continued in production for very long periods.
Products from crafi producers, such as Ferrari and Aston Martin, and models in production
lor more lhan twenty years, such as the Morris Mini and Citroen Deux Cheveux, have been
At the same time, the European volume firms were pursuing
d mass-production strategies as they stmggled to
the companies taken over in the 1970s and 1980s. They
ced their models on offer slighrly, from forty-nine to fortynd allowed remaining models to age rnarkedly. SpeciíiPSA (Peugeot) rationalized the product offerings of Citroen
hrysler Europe while Fiat consolidated the offerings of AIfa
eo. Recently Volkswagen has absorbed the Spanish producer
(which previously built Fiat designs under license), Volvo
Renault have agreed to collaborate on auto-making activities,
General Motors has become the senior partner in a joint
n ure involving Saab's automotive operations. These evenls
ggests that one more round of product rationalization may be
rried through in Europe in the early 1990s.
The Americans by contrast have managed a substantial inir product ranges, from thirty-six to fifty-three modat a cost, as you can see along the bottom axis of Figure
bers here refer to the average number of years the
product has been iil production. In the case of the Japaducers, this number is between 1.5 and 2.0 years-about
at one would expect from companies with policies of repiacing
ry model every four years. For the Americans, by contrast, the
rage age has risen from 2.7 to 4.7 years, suggesting that the
verage model is now kept in production for nearly ten years
an the eight years common in the past. The reason, >ve
simply that the Americans, with their inefficient prodpment processes, are finding they do not have the
engirleers to expand their product range and renew
rth American motor-vehicle market, as
in Figure 5.3, indicates that the Japanese strategy of the
is likely to continue in the 1990s. As of the 199 1 model year,
se companies still offer no products in the full-size car,
n, and pickup truck classes. Similarly, their range of ofirrings
the European luxury and sport specialty classes is stil! very
est, despite the reccnt excitement about Lexus, Infiniti, and
a. Large cars and pickup trucks are the most profitable areas
the entire world market. So it would be remarkable if tbe
roducers did not move ahead rapidly to complete their
roduct offerings in the larger size classes of cars, tnicks, and
ans, and perhaps to develop new market segrrients as well.
At the same time, the European volume producers ~villsoon
124
THE MACHINE THAT CHANGED THE WORLD
2500
-
2250
-
per product. (The years 1955, 1973, and 1986 have been used
cause tliese were years of peak demand in this highly cyclical
rket. To make the 1989 sales-per-product figures comparable
he earlier years, we have assumed that 1989 sales were a t the
86 level when tliey were actually about 9 percent lower.)
+ Arneiican
--B-
Euiopean
1955
SaleslProduct (000s)
Share of Market Captured by 6 Largest-Selling Praducts
18"2
1983
,984
,985
1886
XsBr
,988
issl
1986
19119(2J
30
84 117
259 169 136
73 43 25
142
112
24
25 38 47
309 322 238
50
219
i9lo
Yeai
Note:
1973
"Models" are as defined in figures 5.2 and 5.4. This figure has been estimated by doubiing
the average product age shown in Figure 5.2 and multiplying by annual production voiume
shown in Figure 5.4. Come iype al estimation is unavoidable. because most af the model
counted in the figure will contini!e in production for a number of years.
ropean Products: (3)
30
Source: Calculated by Antony Sheriff from PRS and Aulornobile Reviewdata
11
27
35
27
SaleslProduct (000s)
26
18
5.4) for the Americans, the European volume producers, the European specialists, and the Japanese. Because the Japanese models-with very few exceptions-remain in production only four
years compared with eight to ten years lor the American and
Europeas? volume producers, iL is not surprising that the Japanese
are producing only one qliarter as many copies oleach car during
its production life. What is surprising is that the very long model
O
O
19
55
41
94
58
SalesIProduct (000s)
been happening in a specific market, notably the U.S. car, van,
and truck market. We have already seen in Figure 5.3 that a broad
array of market segments are now served with a surprisingly even
leve1 of sales across SegnlentS. Figure 5.6 shows the dramatic
5
73
Source: Caicuiated by the authors from sales data in Ward's Autornofive Reporls, January 8, 1990
(for 1989) and in the Ward's Autornotive Yearbook(f0r 1955. 1973, and 1986).
THE MACHINE THAT CHANGED THE WORLD
for an assignment to an engineering specialty, perhaps in the
engine department.
At first, thev are likelv to be assiened to a new-~roduc
development team. There they will do very routine work, largel
adapting established designs to the precise needs of the
model. This task, as we saw in the preceding chapter, conti
for up to four years.
After successfully working on a new development project, th
young engineer is likely to be transferred back to the engin
department to do more fundamental work, perhaps on the desi
of a new engine, such as the V6 and V8 units recently introdu
by the Japanese producers and intended for use in a &ole rang
of new models. (An engine-development program, like a ne
model development program, requires three to four years betwe
initial concept and actual production.)
Once the engineers successhilly finish their stint on th
second type of development team, some of the most promising ar
selected for additional academic training and then are set to wor
on longer-term and more advanced projects. For example, th
engineer miglit study how to incorporate fiber reinforcements
into highly stressed metal parts, such as the rods connecting the
crankshaft to the pistons. In working on these projects, the engi
neers consult closely with academic experts on retainer to th
company.
However, even these longer-term development projects have a
very specific objective-to remedy some weakness in the company's products identified by the product or major-componen
development teams. So they are tied tightfy to the needs an
tjmetahle of specific development projects. And, the work is conducted by engineers who thoroughly understand the practicalities
of product development and production. To make sure that the
engineers maintain their sensitivity, Honda, for example, assigns
even its most advanced engineers to spend a month of each year
working in one of the other functional areas of the company-the
selling divisions, factory operations, supply coordination, and so
forth.
The Japanese lean producers exercise extreme care not to
isolate their advanced technologies from the day-to-day ~vorkings
of the company and the incessant demands of the market. Based
on their observations of U.S. and European mass-producers. they
long ago concluded that, to be eFfective, engineering, even of the
most advanced sort, must be tied into the key market-driven
nrtivitir~
nf thr romnanv.
-
ERN lNNOVATlON IN PRACTICE: LOW-TECH WEAKLINGS TU HIGH.T£CH WONoERS
how this process can work is nicely illustrated by the evoluof Japanese engine designs during the 1980s. At the beginning
decade, the Japanese companies faccd a common problem.
y had assumed that energy prices would continue to climb
that consumers would want smaller cars, so they had invested
ns of dollars in the late 1970s in new engine plants for small
ur-cylinder engines. Instead, fue¡ prices fe11 and consumers were
ooking for larger cars with more power.
What to do? Engine sizes can be increased slightly using
xisting production tools by ividening the bore of the cylinders
nd increasing their stroke. However, to go further-by addin::
linders or changing the engine's configuration, from, say, four
ylinders in a line to six in a V arrangement-would be staggergly expensive, because it would require junking most of the
isting production tools. New billion-dollar engine plants would,
turn, drain resources from the product-development teams
raining to rapidly increase the range of Japanese products.
urely, the lean producers thought, there is a quicker and easier
In fact, there was. The product-development teams turned to
the advanced engineering groups, which suggested introducing
every available technical feature to boost the performance of the
basic four-cylinder engines. These features were conceptually
simple-fue1 injectiori rather than carburetors, four valves per
cylinder rather than two (to get more fue1 in and more exhaust
out on each stroke), balance shafts in the bottom of the engine (to
damp the inherent roughness of four-cylinder designs), turbochargers and superchargers (to get more power from the same size
engine), a second set of overhead cams (to make valve timing
more precise), and even an additional set of cams for use a t higher
engine speeds (to get full power kom the engine across a wide
range of operating conditions).
In addition, the engineers worked very hard on what is known
in the industry as refinement-paying attention to the smallest
details of an engine design so that the finished engine runs
smoothly and without complaint at al1 speeds and in al1 driving
conditions, imitating the performance of a much larger engine.
Finally, the engineers paid endless attention to manufacturability. Because they were going against one good engineering
JHE MACHlNE JHAJ CHANGED THE WORL
practice-by adding parts and complexity to an already
device-they had to work extra hard at manufacturabili
complex engines would work properly every time and e
the minimum of extra production expense.
A s these features were added during the decade of the 19
they had an interesting effect on public perceptions
was perhaps unanticipated. Even as they raised the power of
same basic engine, in some cases by a factor of two, these in
tions convinced buyers, particularly in North America, that
nese cars were now "high-tech," that they now had the
advanced features. They had grown from "low-tecli" weakli
1980 to "high-tech" wonders by 1990 while preserving their
ufacturers' basic investment in production facilities for
engines.
This perception on the part of consumers was enormousl
frustrating to engineers in many of the mass-production c
nies wlio knew that al1 these "innovations" had been
motor industry for decades. For example, four valves per cylinder
and double overhead camshafts were available on the 1924 Bentley, and superchargers were common on the largest European
luxury cars in the 1930s. However, they had often been vetoed by
management as too expensive or complex for production use o
restricted to use in a limited range of specialty models.
What's more, when the mass-producers, particularly in the
Unitcd States, tried to copy these "innovations" on a wid
the weaknesses of their engineering systems were exp
many cases it took years to introduce a comparable featurc, an
often it was accompanied by drivability problems or high
tion costs. GM, for example, Iagged Toyota by four years i
introducing many of the features we just listed in its Quad Fou
engine; it needed two more years to reach a high leve1 of refine
ment. Even then, manufacturing bottlenecks meant that this en
gine was available in only a narrow range of GM cars using four
cylinder engines.
LEAN PRODUCTION VERSUS MASS I N RESEARCH RND DEVELOPMENT:
SOME NUMERICAL COMPARISONS
Given the contrasting approaches to innovation, it's not surprising that the examples we cited are typical and that performance
in the development of new technologies differs systematically. In
m00-
aooo-
Note:
-
Arneiican
Eurcpean
--o- Japanese
-h-
The figures are for worldwide spending on research and deveiopment by firrns in !he motor
vehicle industry grouped by headquarters region. Thus Generai Motors' worldwide spending
is consolidated under "American" and Volkswagen's woridwide spending is grouped under
"European."
Figures are constan! 1988 doliars at 1988 exchange rates.
THE MACHlNE THAT CHANGED THE WORLD
FIGURE 5.9
A NEED EORR EPOGHAL INNOVATIONS?
Molor Vehicle lndustry Patenting, 1969-1986
-?s-
Arnsrican
European
Japanese
Year
Note:
Figures are lor patents granted by the U.S. Patent Office to assernbler and supplier lirm
located in each main region. In case of subsidiaries whose paren! is headquartered in on
iegion but which operate in another region. the patents were counted in !he region
operation. For example, Alired Teves is a German subsidiary of !he U.S.-headquartered I
Teves' patents have been counted in the European region.
Patenting by supplier lirms was estirnated by developing a list o i majar autornotive supplier
headquartered in the three principal regions, using the following sources:
Japan: Dodwell Consultants, The Structure of the Japanese Auloparls lndustry, ~ o k y
Oodwell, 1986
North Arnerica: Elrn Internalionai. The E h Guide to Automotive sourcing, 1987-88, Ea
Lansing. Michigan: Elrn lnternational. 1987
Europe: PRS. The European Automotlve Components Industry 1986, London: PRS. 1986
This list was then cornpared with data on patents by cornpany, provided by the U.S. Oltice
of Technoiogy Assessmenl. Adiustrnents were rnade to exciude nonautomotive patenting by
large multi-product firrns, such as Allied Signal in the United States and Hitachi in Japan.
far we've talked about innovations that involve the introducn in production vehicles of ideas already Fairly well understood
nical level. We've listed a number of advances of this
in the 1980s, and many more will be available in the 1990srticular, the application of electronics to mecfianical vehiele
h as vehicle suspension and the availability of mobile
unicaiions at lower cost in a much wider variety ol vehicles.
what about epochal innovations-really big leaps in teclinoal know-how such as would be entailed in workable fuel-cell
r units or all-plastic body structures or sophisticated naviand congestion-avoidance systems? As we will see, the
may prove a time for such innovations. Can lean producers
spond to these much more daunting challenges?
In fact, the world auto industry has lived during its first
ntury in a benign environment-demand for its products has
sed continually, even in the most developed countries; space
s been available in most areas to expand road networks greatly;
d the earth's atmosphere has been able io tolerate ever-growing
motor vehicles, with minor technical fixes in the 1970s and
designed to solve smog problems in congested urban areas.
ortly, the environment for operating motor vehicles may bemuch more demanding.
Demand lor cars is now close to saturation in North America,
, and tlie western half of Europe. A small amount ol increental growth will be possible in the 1990s, but by the end of the
producers in these markets %vil]need to provide consumsomething new if they want to increase their sales volume
in dollars or marks or yen rather then units). Moreover,
th ol vehicle use and increasing resistance to road buildhave made the road systems of these regions steadily more
gested, gradually stripping motor-vehicle use of its plcasure.
New electronic-vehicle technologies that permit vehicles to
around congestion and even, some day, to drive themight solve both problems: Handing driving over to the
ould permit car companies to charge much more per
they did not se11 more units-and in-vehiclc entertaintems could be moneymakers as well if drivers were
elieved of the need to watch the road.
THE MACHINE THAT CHANGED THE WORLD
Meanwhile, in the 1990s cars and trucks that are able t
gather information from the roadway about congesti
find the fastest route to their destination could make m
use of Iimited highway space. Given the size of the potential pr
it's no surprise that governments and motor-vehicle compani
North America, Western Europe, and Japan have recently initi
publicly funded cooperative research programs in each regi
find technical solutions to these problems."
However, making these technologies a reaiity is trul
ing. The computer industry is still a long way from a
computing power Sor autopilots, and the reliability of suc
tems would have to be very high. Although motor vehicles
human control kill more than 100,000 people each year in
America, Western Europe, and Japan combined, it's hard to ima
ine that the public would accept a computer-controlled syste
that killed a half or even a tenth as many people.
What's more, solutions will have to be sought lar beyond the
research labs of individual companies, both because publicly
owned roads will be a key element in the necessary information
systems and because the standards selected may have a major
bearing on the health of national motor industries. The recent
debate over worldwide standards for high-definition televisionin which governments in each major region have jockeyed to
provide an advantage for their home team-are perhaps a foretaste of what is to come in the motor-vehicle industry.
A breakthrough in navigation and autopiloting could revitalize the consumer's desire to spend discretionary income on
vehicles, even in the most saturated markets. However, even m
startling breakthroughs in motor-vehicle technology may
needed simply to preserve what society has already, if the wor
predictions about the greenhouse effect are borne out. These foc
on the potential effects of the rising levels of carbon dioxi
(partly from motor vehicles), methane, and chlorofluorocarbo
(partly fr-om auto air conditioners) in the earth's atmosphe
These emissions may dramatically raise temperatures and alte
the global climate if they're allowed to continue.
In the worst case, early in the next century we ma
dramatic rise in sea levels as the Antarctic ice melts,
much of the world's eoastal plains where population is
trated. We may also see rairifall patterns change to con
world's breadbaskets to dustbowls. Even much more mode
ESIGNING THE CAR
137
eaten the earth's ability to support its current
Currently, the scientific debate on the greenhouse effect is
raordinarily confused. Everyone agrecs that carbon dioxide,
thane, and chlorofluorocarbon levels are rising, but the precise
sequences of the increase are far from clear. Computer models
orporating the many feedback loops in the climate system are
keys to prediction. So far, however, the models are in only
ent and make their predictions only within a broad
ge. What's more, the fate of specific regions as climate changes
On the other hand, society is now pouring enormous scientific
ding precise answers, probably within the next
years. We would be surprised if the motor-vehicle industry
es not have to respond in dramatic ways-and its response may
est of lean approaches to research and develople, in tlie extreme case, emissions of carbon
xide might have to be eliminated altogether, creating the need
hydrogen-powered cars, which produce only water as an end
duct of combustion, or even solar-poruered vehicles.
So far, the Japanese lean producers have not lailed at epochal
ovations of this sort; they simply haven't tried, occupying
mselves instead with a brilliant scavenging process that has
nned the technological landscape for ideas nearly ready for the
ket, as in the case of the high-tech four-cylirider engines of the
Os. A much more difficult challenge probably lies just ahead.
e could impose the cost and efficiency discipline of the market
hile preserving the coordination advantages oE a unified
Sloan had a solution to the problem of the cvclical car market
s well: When the marlcet slumps, lay off workers in the supply
l
COMIDINATINI; THE SUPPLY
CHAlN
u
The modern car is almosi unimaginably complicated. As we noted, a typical model is made up
of more than 10,000 parts, each of which must
be designed and made by someone. Organizing
this enormous task is probablv the greatest challenee in manufacturing a motor vehicle. Yet it is the one least understood and
appreciated by the outside world.
Henry Ford thought he had solved the problem by the time of
World War 1: Do it al1 yourself in your own company. However,
his solution raised as many questions as it answcred: How do you
organize and coordinate hundreds of thousands of employees in
hundreds of factories and engineering offices? What do you do
with your machines and factories, al1 dedicated to making specific
parts for your own products, when demand shifts or tlie economy
goes sour?
In the 1920s, Alfi-ed Sloan found one answer to these problems: Do it al1 in your own company, but set up decentralized
parts-making divisions as independent profit centers-For example, Harrison Radiator, Saginaw Steering, AC Spark Plug-to
make specific categories of parts Eor the whole company. By
treating the divisions as independent businesses, Sloan thought,
stem just as you lay off workers in the assembly plani.
Bv the 1950s. the Ford Motor Company
. ., under Henrv Ford 11
a new idea, which, as it turned out, was actually an uld idea.
rd put out to bid to completely independent supplicr firms
any categories of components Eormerly supplied Erom ~vithin
e company. The suppliers were given detailed drawings oC the
ecessary parts and asked for their price per part. The lowest
der generallv won a one-vear contract. When ihe market
umped, these suppliers were laid off by means of canceled
tracts, just like workers. This was, in fact, the world Ford had
around 1913; the ~vorldof arm's-length, market-based, shortrm interactions with independent businesses.
In the 1980s mass-production companies around the world
ere using both approaches. GM was the most integrated, with
out 70 percent of the parts in each car and truck supplied by its
-house parts divisions. Saab, at the opposite extreme, made
nly about 25 percent of its parts, always preserving for itself just
hose parts most visible to the consumer-the body and the
u
This cliapter is based primarily un the research ol Toshihiro Fujimoto and
Richard Lamming.
How much each company actually integrated was a f u ~ ~ c t i o n
he company's history and its size. The enormous investment
had sunk into its parts operations made it very hard to ihink
out alternatives, while Saab was simply too small to makc al1
own parts. (Indeed, a key justification of the Ford purchase of
aguar and of the GM joint venture with Saab has becn thai
guar and Saab can obtain cheaper parts through the greater
argaining power of a larger producer and can share some comon parts, such as switches and lights, with Ford and GM,
spectively.) However, neither system-in-housc or arm'sength-works very well.
In the mid-1980s, in the twilight of mass production, many
ompanies, including General Motors and Chrysler, experimented
th reducing the fraction of parts they obtained from their inuse suppliers. This tactic was inspired by a belief that lower
ages in outside supplier companies were the competitive secret
f the Japanese supply systems.
In our view, this change in direction-now on hold a t Chrysler
THE MACHlNE THAT CHANGED THE WORL
and GM because oS resistance by middle management and the
UAW-largely misses the point.' The key to a competitive par
supply system is the way the assembler (for example, Ford
Renault or Toyota) works with its suppliers (Sor example, For
Automotive Trim Operations or Benctix, Renault's Transniissio
Division or Valeo, Toyota's Engine Division or Nippondenso).
Wbether the supplier comes Srom inside the company o r o
makes surprisingly little difference. To see why this is so, le
pickup where we left off in Chapter 5 and follow tbe component
supply process as it has worked (and in many cases still works)
mass-produciion car cornpanies. We'll trace the system from t
moment it begins in the design of a new car.
1
1
HATUFE HASS PRIBlüN: DESIGNING TliF PNTS
Remember that the design process in a mass-production compan
proceeds in a sequence, one step at a time. First, tlie overa
concept for the new model is specified by the assembler's produ
planning team and reviewed by seriior management. Then t
product is pianned in detail, down to the fractiori of an inch
example, the wheelbase and track) and the specific type os m
rial to be used for each pari (for example, steel fenders, plast
steering wheel, aluminum enginej. Next, detailed engineeri
draxvings are made for each part, specifying the precise materi
to be used (steel of a given gauge with double galvaniied coati
for the fenders, for example; thermosei plastic wiih carbon fib
reinforcement for the steering wheel; specific aluminum allo
the engine block, and so forth). Only at this point are tlie organ
zations that will actually make the parts called in. These typicall
number between 1,000 and 2,500 lor the complete car, includin
independent companies anci in-house parts divisions.
When the suppliers-wlieilier in-house or independentnally get the call, they are shown the drawings and asked Sor bi
For example, they're asked, "What will the cost per steering wht
be Sor 400,000 steeririg wheels per year?" The mass-producti
assembler also sets a quality target, so many bad parts per 1,0
as the acceptable upper limit, and a delivery schedule-perha
one or two tinies a week-with a penalty for failure tu dcliver
time or in the right quantity. The term of the contract is genera
quite short-usually a year Sor parts that require new capita
ORDlNATlNG THE SUPPLY CHAIN
vestment, but even less for those commodity parts, such as
teries or tires, that most companies in the industry buy from
same suppliers and that are already in produetion. So price,
ality, delivery reliability, and contract length become the four
elements of the assembler-supplier relationship.
When the suppliers see the drawings, they know from long
perience that they are involved in a complex game, where none
the real rules are written into the bid tender. They realize that
e assembler's procurement office is under treniendous pressul-e
duce costs. "Cost comes first" is the assembler's byword. So
ing a low priee per part is absolutely essential to a winning
d. However, the suppliers are also aware that follow-on business
a new model can often extend for ten years. Then there's the
arket for replacemerit parts, which may be eonsiderably longer.
o, in reality, they are not bidding on a one-year contract but,
potentially, on a stream of business running for twenty years.
Since this is the case, should they bid below cost? Doing so is
empting, because the suppliers' experience also tells them that
ce a part is in production, with aeceptable quality and delivery
rformanee, they may be able to go baek to the assembler for a
st adjustment: "We can't get our steel in the net shape we need,
scrap costs are running above our estimates," they may say
eaning, that the only steel blanks they can get are too big, so
y must eut off more to get the size of part they need, a process
tails extra initial cost and waste), or "Our union is insisting
k-rules changes that are increasing our costs" or "The new
e we bought to mold steering wheels cannot provide adeate quality without hand finishing."
In addition, there exists a long tradition of the annual cost
ustment for follow-on conti-acts, designed lo allow for general
ation. The assembler tends to grant these adjustments across
e board without investigating individual circumstances. To look
nto each case would siiiiply require too much effort. Of course.
ppliers are almost eertain actually to reduce production costs
er time, since, as time goes on, they gain experience in producthe part. So the aniiual cost increase in subsequent years may
rn a money-losing initial bid into moneymaking follow-on
Finally, for some parts requiring heavy investments in new
oduction tools, the assembler may find i t extremely costly and
convenient to obtain a new supplier once production is in full
ing. Supplier-s of these parts may gamble that their ability to
raise prices will grow over time. This mind-set makes the tem
tation to "buy the business"-that is, put in a deliberately low
bid in order to get a foot in the door-almost irresistible.
The mass-production assembler has played this game tho
sands of times and fully expects the successful bidders to co
back later for price adjustments. Thus it is important for
assembler's product designen to have some idea of suppli
real costs, so they can accurately estimate price adjustme
downstream.
It's hard work, though. A key feature of market-based biddi
is that suppliers share only a single piece of information wi
assembler: the bid price per part. Othenvise, suppliers jeal
guard information about their operations, even when they a
divisions of the assembler company. By holding back informati
on how they plan to make the part and on their interna1 efficienc
they believe they are maximizing their ability to hide profits fro
the assembler.
Once the assembler designates the winning bidders, the su
pliers set to work making prototype parts. This process is like
to uncover many problems, because the traditional mass-pr
ducer farms out the many parts in a complex component toma
suppliers who may have no direct contact with each other.
example, until recently General Motors built practically al1
own seats by ordering about twenty-five parts per seat from
many suppliers. When the parts were finally put together in t
finished seat, it was not surprising that a piece wouldn't fit or ih
two abutting materials would prove incompatible. For examp
they might rattle or squeak in cold weather because of differen
expansion coefficients.
Once the supplier tests the parts in components and th
assembler, in turn, tests the components in complete vehicles, th
assembler specifies necessaq changes in each part and gives th
sign-off to begin volume production. However, the mass-prod
tion assembler is still not through with the supplier-selecti
process.
1
I
osts among suppliers whose operations it only vaguely undernds. The obvious way to do sois to identify additional suppliers
each part and give them the final, production-ready dra\vings
the basis for making bids. The suppliers who already have been
ected are horrified, of course-which is precisely the idea. The
'tial supplier also feels cheated, because the new bidder will not
ve to bear the cost of fine-tuning the original drawiiigs.
Of course, the first supplier has also piayed this particular
me countless times before and has probably left some room in
bid to make adjustments in subsequent years, as the assembler
two or even three or four suppliers against each other.
'S more, many of the assembler's threats to seek an alternae source may turn out empty, particularly when they are
ected at in-house suppliers.
Let's take the example of one o€ GM's in-house suppliers.
1 imagine that the program manager for a new GM product is
ppy with the in-house supplier's bid-it's too high and, in
e past, the supplier had quality and delivery problems. Hower, no sooner does the manager identify an alternative bidder
tside the company than the in-house supplier goes to corporate
adquarters and explains that loss o[ business 011 this part will
quire an increase in the cost OS similar parts already being
pplied for other GM products. Why? Because economies of scale
ill be lost and the in-house supplier will Iiave excess capacity.
Headquarters, always very respectful oF scale-economy and
pacity-utilization justifications in a mass-production firm such
GM, then has a talk with the program manager. The in-llouse
pplier makes solemn promises to try harder to reduce costs in
future while improving quality and delivery reliability-and
S the business. In this way, the interna1 market, which supposly keeps the in-house supply divisions honest, is gradually
uted. This process explains how GM managed to have both the
rld's highest production volume and the world's highest costs
many of its components supply divisions through much o€ the
At the end of the selection process, the assembler usually ends
with a single s u.o.~ l i e rfor the most complex and technologiIly advanced components, such as engine computers. For comodity parts such as tires, three or four suppliers are often put
der contract. However, designating the coinplete roster of supiers and beginning volume production is only ihe end oZ the
st stage of assembler-supplier collaboration on a new product.
v
MlllUíít MhSS
SUPPLYING THE PARTS
At this point, the purchasing department is worrying less abou
getting the vehicle into production and more about how to contro
JHE MACHlNE JHAJCHANGED JHE WO
159
This development occurred at the same time that man
semblers began to outsource the production of parts t
produced more economically by specialist suppliers th
house divisions. ("Outsource," in car-industry jargon,
means buying a part from another company rather than
it yourself.) In the early 1980s, Sor instance, Ford in the
States shut down its in-house wiring-harness assembly opera
and gave this business to twelve outside suppliers. Later i
decade it reduced the number of suppliers to f o ~ r . ~ '
Assemblers can reduce the number of suppliers in three
First, they can tier suppliers by assigning whole co
to a first-tier supplier-seats, for example-as the Japane
This tack can reduce the number of suppliers from twentyonly one, as we saw earlier. The assembler's administrative
for coordinating supply plummets.
Second, even without tiering, assemblers can cut the nu
of suppliers by reducing the parts count in component
ter 4, we showed how front-bumper assemblies in a GM
contained ten times the parts of a similar assembly in a Ford
So GM might also have ten times as many suppliers to its ass
bly plant. Because cars and trucks are becoming more co
cated, partly because of environmental demands and par
satisfy consumers, there will always be a race between a gro
number of vehicle systems and a declining number of part
system. For the moment, however, parts counts are falling fa
with the result that assemblers are reducing the number of
pliers.
Third, assemblers can single-source parts that previousl
two or three suppliers. They can convert to single-sourcin
a traditional market context by soliciting bids, then giving
business to the low bidder. The supplier who gets the
business should then have greater economies of scale and,
lower prices. Respondents to our mail survey confir
single-sourcing was indeed a trend. On average, between 1983
1988, the number of suppliers producing a specific part for
American assembler fell from 2 to 1.S and the number of
producing the same general type of part for each ass
from 2.3 to 1.9.14(As shown in Figure 6.1, this is in sharp cont
with Japan, where multiple sourcing is the rule.)
The main reason assemblers go to single-sourcing is
longer production runs of a single component and av
tion of tooling. There's a down side to single-sourcin
the assembler vulrier-able to supply disruptions,
ng recent strikes affecting Ford and Renault in
though many observers have argued that single-sourcing is
er useful technique Western assemblers can learn from the
se, we've already seen that this argument is both wrong
int. These same observers assumed that singleg in Japan led to longer-term relationships with suppliers.
,as we saw, Japanese long-term relationships do not depend
but on a contract Eramework that encourages
nother change in the Western supply system is the shifting
uality among assemblers. Al1 the U.S. assemed a quality grading system for suppliers-not
a shipment-by-shipment basis but for al1 parts supplied
le period. Ford started a systematic supplier
system, called Q1, in the mid-1980s. QI was followed
e GM Spear program and Chrysler Pentastar. These
ical systems that rank suppliers by the number
red in the assembly plant, delivery perfores, progress in implementing quality, improvement proin the supplier plant, leve1 of technology, management
s, and more. The aim was to bring every supplier graduto higher and higher levels of performance and quality.
d a major impact in diffusing quality-monitores, such as statistical process control (SPC), to supC, tool operators record the dimensions of each
parts-produced. If they notice these dimenwhat they should be, they either make the
nts to the machine, or, if it's a more difficult
as a machine malfunction, cal1 for help. In theory,
tive part should be produced. Part of the Ql program
going one step further and sharing these SPC charts with
survey showed that 93 percent of suppliers used
r operations in 1988, up from 19 percent in 1983.25
nese began improving quality in their suppliers through
process, though they diffused SPC to their suppliers in
some thirty years ago. Obviously, mass-producers
way to go. In fact, once suppliers use SPC systems
Sor a while to identi- when a machine is about to produ
defective parts, find out why, and then take steps to ensure t
prohlems don't recur, SPC becomes a routine actisity Sor prod
tion workers, a point many Japanese companies reached in
mid-1960s.
The next step in the path to lean supply, of course, would
the sharing oS detailed infoi-ination on the cost of each producti
step, using value-analvsis techniques. Ironically, General Elec
originally developed these techniques i n 1947, and they w
enthusiasticaily adopted by the Japanese in the early 1960
1988, however, only 19 percent o l U.S. suppliers were pr
this kind of information io tlieir assembler customers. T
shouldn't come as a surprise, since no fundamental change
curred in the adversaria1 po~ver-basedrelationship between
semblers and suppliers.
We have seen some shift toward more kequent deliveries
the West. Iii 1983, more than 70 percent of U.S. suppliers delivere
more than a week's supply of parts a t once (that is, they deliver
once a week or less frequently). Today, this number has fallen
20 p e r ~ e n t This
. ~ ~ percentage compares with 16 percent of Jap
nese suppliers delivering weekly as Far back as 1Y82.2Thatye
52 percent of Japanese suppliers were delivering daily and
further 31 percent Iioiirly. tn the United States, only 10 percent o
suppliers were delivering daily or hourly combined by 1988.
The American improvement in delivery schedules, though
not a move toward lean supply. Rather, it is an attempt to cut
amount oS inventory in the assenibler's plant; the supplier,
stead, keeps the inventories. So the change doesn't represent
philosophical shift but simply an attempt by assemblers to s
costs to their suppliers.
Moreover, it is one thing to deliver smaller lots of parts
frequently to the assembler, but quite another to produce
parts in smaller lots, as a lean supplier would do. In faet,
percent of U.S. suppliers produced more than a week's supply
a part at one time before changing the tools to make another pa
hardly any change from 60 percent five years earlier.2'
Many suppliers iii ihe survey still express skepticism ab
the just-in-time concept. That they do is, perhaps, not sui-pris
given the way the concept has been used by assemblers thus f
The suppliers, with some justice, see just-in-time as a wa
shifting tlie burden of inventories onto them. Part of the prob
is that, initially, just-in-time was thought of as frequent delive
the assembly plant. Hoivever, as we saiv in Chapter 4, jusi-inme comes into its own only when i t is applied to production.
e discipline imposed in the plant by manufacturing small lots
one of the key steps to greater efficiency and quality in lean
another sign that conditions aren't changing radically, our
survey revealed no eiiidence tu suggest. that U.S. suppliers
ght U.S. assemblers any more trustivorthy than they did five
we did see some moves toward longer-term
rs ag-although
tracts. The average contract length rose lrom 1.2 years to 2.3
rs, and the proportion of suppliers with contracts oí' three
ars or more rose lrom 14 percent to 40 percent.j" At the same
e, suppliers reported that the assemblers had given them little
sistance in reducing costs and adopiing neiv techniques-a
'ng that supports our impression that those relationships are
istant as ever.
It is true that supplier engineering, combined with long-term
traets (three to five years instead of a year or less), higher
ality standards, more Srequent deliveries, and single-sourcing
many components characterize a n e i ~North American supply
stem of the early 1990s. Don't be Fooled by these developmeilts,
ever, into thinking that Westei-n suppliers have been rnoving
ward lean supply. They have not. While many of the changes
semble what Japanese lean supply looks like from thc West,
arly al1 have been driven by cost pressures and existing massoduction logic: single-sourcing lar achieving economies ofscale,
st-in-time for shifting the burden of inventories, and more.
Indeed, without a fundaniental shift away from a poiver-based
gaining 1-elationship, it is almost impossible to move toward
n supply. If the assemblers don't esiablish a new set of ground
les for joint cost analysis, price deterrnination, and profir sharg, the suppliers will continue to play by the old rules.
Confronted with this power-based relationship, the suppliers'
in objective is to shift any advantages to their side. Their chief
ay of doing so has been to introduce neiv technologies and bring
gether discrete components into systems. Without detailed
ue analysis, the assembler is unable to do more ihan guess the
ce OS a complex component or to play ofS one supplier against
The incorporation of niany necv technologies into the autoobile, such as antilock brakes, elecrronic engine-managemerit
stems, and plastic body parts, is giving some suppliet-s a greater
role in designing not only discrete parts but whole system
also bringing many new suppliers-giants such as Motoro
mens, and General Electric Plastics-into the industry for t
time. The more complex the technology, the less it fits
tional mass-production supply systems where the asse
the upper hand. Companies supplying technologically a
or complex components have an opportunity to add mo
or, in other words, improve their bargaining power vis-a
assemblers. For many suppliers, this has been the prime
tor for moving to more advanced technologies.
l
While al1 these changes in the relationship between suppl'
assemblers have been occurring in the United States, w
been happening to supplier manufacturing performance?
is the gap between U.S., European, and Japanese supp
answer this question, Toshihiro Nisliiguchi conducted a su
fifty-four matched components plants in Japan, Europe
North America.3' The results, as summarized in Figure 6.1
that manufacturing performance among Western comp
companies has been no better than that of the assembfe
other words, the performance gap we found when we com
assembly plants is mirrored in the supplier industry.
In terms of component quality, the United States was
shouting distance of the Japanese-with 33 component
per 100 cars compared with 24 for the Japanese. The Eur
at 62 defects per 100 cars, still lagged far behind. On a
measures, however, such as the time it took to change
dies, the leve1 of inventories, the number of job classificat
plants, the degree of multi-skill working, and delivery fre
Nishiguchi found a significant gap between parts maker
United States and Europe and those in Japan (see Figure
most cases, the gap was larger than in assembly-a fac
indicates that the components industry is some way behin
assemblers in adopting lean manufacturing.
However, al1 is not lost, as there are now at least 145 Jap
components suppliers Iocated in North America, and many
U.S. suppliers are beginning to supply the Japanese transpl
ed States. Those U.S. suppliers who llave managed to
tracts with the transplants have an excellent opportuearn everything from lean manufacturing and product
pment to lean supplier relationships. And it can be done.
hen, for example, GM's Packard Electric Division began to
the GM-Toyota joint-venture NUMMI plant in California,
al shipments of wiring harnesses were judged competitive
e but not on q ~ a l i t yAfter
. ~ ~ discussing the situation with
1, Pacltai-d stationed a resident engineer a! NUMMI fullo that quality problems eould be attacked immediately. It
ought
technical assistance (in the form ol three industrial
ers loaned for six months) from Sumitomo Wiring Systems,
Toyota's traditional suppliers of wiring harnesses. The
heloed Packard install the full Tovota Promo enrineers
n System in its plant at Juarez, Mexico, dedicated to supplyUMMI. The results of Packard's continuous efforts to learn:
eighteen months it advanced from the bottom to the top in
MI'S supplier rating.
there is still ample potential for misunderstanding the
ces in supplier philosophies, as the Collowing example
'
U.S. supplier of a coniplex part won business from NUMMl
steadily impi-oved its defect ratc 2nd delivery reliability to
perfection. Then, it put in for a large price increase, a move
seemed reasonable by Western standards, since it had proved
pabilities. However, to NUMMI's Toyota-trained purchasing
the request seemed a blatant example of bad faith. In the
system, suppliers should never commit themselws to dcg at unrealistic prices but must be prepared instead to
their price continually over the life of the model. This kind
sunderstanding illustrates the difference in approach bemass and lean supplier relations that has yet to be bridged.
TfRN EURDPE AS A HALFWAY STATION
supplicr system changes in North America, it is coming lo
ble not so much the supplier system in Japan as in Western
e. Although the mass-production assemblers in Western
e, as we saw in Chapter 4, are now the world's most ortho-
dox Collowers of Henry Ford in their own factories, the Europ
supply system has always differed from mass-production meth
and has been somewhat closer to lean supply.i4
That's partly because European assemblers have always b
smaller and more numerous. Six companies divide the
market with shares of 10 to 15 percent of total production, w
a half-dozen specialist companies split the rest of the ma
These are shown in Figure 6.2 for the 1989 sales year.
These smaller assemblers never had the scale or the fund
contemplate doing everything themselves, as Heriry Ford
initially and GM very nearly did for fifty years. What's more, t
have always existed a number o£ strong European suppliers
by the German firm Bosch, but including GKN (universal joi
and SKF (bearirigs)-with a clear technical lead in certain co
ponent areas. So the tradition in Europe has always been [or
large suppliers to be more talented. Rather than working
drawings, many have engineered complete components for
assemblers. Clark and Fujimoto found, for example, that wh
fact that the European components market is ihe largest in
world and the top twenty companies account for onc-third of
upplicrs around their home-country assembiers, both phvsily and in terms of long-term rclationships. The French assemcentrated in the Paris area, with whom they havc worked for
ents. This developrnerlt is occurring just as Europe itself is
truly regional system, so the leve1 of structural reorganization
Automotive Market Shares in Western Europe, 1989
Praducer
Volkswagen (Audi, Seat)
Fiat (Lancia, Alía Romeo)
Peugeot (Citroen)
Ford
General Motors
(opel, Vauxhall)
Renault
Mercedes Benz
Rover
BMW
Volvo
Japanese
TOTAL
Marhef Share (9;)
15.0
14.8
12.7
11.6
11.0
10.4
3.2
3.1
2.8
2.0
10.9
100.0
mated Number of Components Suppliers in
th America and Western Europe
Majar
1,000
450
400
300
250
50
50
1,500
Minar
4,000
5,000
1,500
1,500
1.000
500
500
10,000
168
THE MACHlNE THAT CHANGED
based bargaining and substituting an agreed-upon rational stm
ture for jointly analyzing costs, determining prices, and shari
profits, adversaria1 relationships give way to cooperative o
Cooperation does not mean a cozy relaxed atmosphere-far h.
it. As we sa%v,Japanese suppliers face constant pressure to i
prove their performance, both through constant compai-ison
they also have much greater discretion than in the west, wi
Feater responsibility for designing and engineering their ow
products.
the end of Chapter 4, we made the important distincti
between the mind-numbing tension of mass-production work a
[he creative challenge of constant irnprovement in lean prod
tion. Very much the same contrast exists in the supplier syste
In m a s production, suppliers are constantly fmstrated as t
trY to guess the assembler's next move. In lean production,
p1iei-s don't constantly have to loolí over tbeir shoulders, Inste
they can get on with the job of improving their own operationswith the knowledge that they will be fairly rewarded for doing
How can the Western post-mass-production supply syst
move toward true lean supply? We suspect that the key mean
will be the creation of lean-supply systems in the west by
Japanese producers, a topic we'll return to in Chapter 9. T
Japanese move wiii force the Western assemblers and their su
piicrs to go the final mile.
OEALlNG WlTH CUSTOMERS
We'\7e now walked through the steps in the motor.vehicle production process-the factory, reand product development, and ComPonents s ~ p p l y In
. each of these areas, we found a
the
methods
and resuits of m a s production
rge gap
d those of lean production. The last stop in our journey takes
to the real reason for these production efforts: the collsumer.
take a look a t how the production system knows what the
customer wants and how he or she goes about buying and maintaining an automobile. We also examine how the manufacturer
goes &out delivering the car to the customer.
why didn2t begin our odyssey with the link between the
customer and the production system? This might seem the logical
place to start in understanding any market-driven manufacturing
process.
the reason: Throughout this volume, we've examined
step of the production process by beginning with the
perspective of mass production. And, as we've shown in earlier
&pters, the success of mass production has been so F a r e d
the manufacturing and design processes that
the needs
customer has tended to come last. So that's the sequence we'x'e
followed as well.
This chaptcr is haced
the research of Daniel Jones, Jan Hclling. and f(oichi
169
172
Car Sales per Dealer in the United States
Producer
1956
1965
General Motors
183
Ford
189
104
351
31 8
21 3
Chrycler
Honda
Toyota
Nissan
173
THE MACHlNE THAT CHANGED THE WORL
Hyundai
Volkswagen
Volvo
nwanted models. This option was much more acceptable to the
ealers but considerably more expensive for the company.
So make matters worse, coordination between the sales divion and product planners in the big mass-production companies
poor. While the product planners conduct endless focus groups
d clinics at the beginning of the product-development process
gauge consumer reaction to their proposed new models, they
ven't iound a way to incorporate coiltinuous feedback from the
es division and the dealers. In fact, the dealers have almost no
k with the sales and marketing divisions, which are respoilsible
moving the metal. The dealer's skills lie in persuasion and
gotiation, not in feeding back information to the product
It's sobering to remember that no one employed by a car
mpany has to buy a car from a dealer (they buy in-house
rough the company instead, or even receive a Sree c a r a s part os
eir compensation ~ackage).Thus, they have no direct link to
her the buying experience or the customer. Moreover, a dealer
as little incentive to share any information on customers with
he manufacturer. The dealer's attitude is, what happens in my
howroom is my business. (In this respect, the relationship beween dealer and manufacturer resembles the one between comnents supplier and assembler.)
As it happened, \ve visited one divisional manager in Detroit
the day he was first shown the prodiiction-ready version of a
ajor new nod del. The manager told us that the car he saw was
tirely different in character and consumer appeal from the
ototype he had agreed to sell two years earlier. Since then, the
les division had had little contact with the product-developent team, which made many changes to the vehicle in order to
ake it easier to manufacture. As far as the sales manager was
ncerned, hcwever, these changes jeopardized the sales appeal oí
car in tlie market, and now it was too late to make any
ustments. In the end, the divisional manager's judgment
oved right-the product has been a disaster.
In Sact, the assemblers' sales divisions havc grown into enorous bureaucracies that cannot elfectively communicate market
mand back to product planners. Moreover, they antagonize the
,alers, with whom they should have a collaborative relationship.
What's more, the bazaar tradition of car selling-where the
ustomer and the dealer try to outwit each other on price-is still
rmly in place at dealerships, even though more and more buyers
~
In some ways, the dealership system has even moved backward since Ford's day. Every assembler maintains an enormous
marketing division lor each of its sales divisions (for example
Chevrolet, Mercury, Dodge) with a main office near headquarters
and regional offices to oversee dealers in each geugraphic are
The marketing division and the dealers typically have strained
relationships, because the division sees its job as making sure
that the dealers se11 enough of the cars the factory must make to
maintain steady production. The sales division's key activity is to
juggle incentives for both consumers and dealers so al1 the cars
are sold.
To achieve this goal, the marketing division inay tie dealers'
orders Sor popular cars to their acceptance of unpopular cars-a
highly effective but extremely unpopular method of meshing
supply with demand. For example, we recently visited a divisional
headquarters at a U.S. company. Headquarters was Sacing the
problem of how to se11 10,000 already built cars that no dealer
wanted. The company had built the automobiles based on its
lorecasts of market demand rather than on actual orders from
dealers or consumers. The market had changed, however, and no
one wanted the cars.
One possible solution was the one we just discusscd: to tie
orders from the dealers for more popular models to the slow
sellers. So to get five popular models, each dealer had to accep
one unwanted model. Another was to offer a Sactory rebate on the
~
rePort in surveys that they thoroughly dislike it. And t
the flow of inforrnation between custorner and dealer is restnct
as well.
Salespeople. that is, aren't really interested in the tus
needs or desires. They want to close the deal as soon
and will present only selected bits of informalion
product to achieve that end. Once the deal is signed, the sales
son has no further interest in the customcr. The entire selling
negotiating system is based on giving the customer as littfe
information as possible-the samc principie on which the
tionship between dealers and manufacturers is based.
The result? As the factory and engineering shop have beca
more efficient under the pressure of lean cornpetitors, the af
the-factory component of the car business-a component t
includes not just dealing costs (manufacturer's
Promotions, shipping, staff and overheads, and more) b
dealer's advertising and warranty work-has accounted
larger and larger fraction of the total cost the consumer
Most analysts currentiy estimate that 15 percent of the buy
total cost is incurred after the factory gate, when the new
turned over Lo the assembler's selling division belore bei
on to the dealer.
As post-factory costs have risen to a larger percent of to
costs, the assemblers have understandably begun to focus m.
attention on pushing these costs do~vn.However, studies OS ret
ing more generally show that auto distribution costs i
America and Europe are already considerably Iower, as a fracti
of total cost, than for many products, including food. ~n fact,wh
the distribution system already offers is low cost, but
even lower leve1 of service.5
We can see other elements of Henry Ford's dealership sys
still firrnly in place today. In the 1980s, the elimination of spe
ordering became a favorite method for m a ~ s - ~ r o to
d utry~ ~ ~ ~
improve efficiency in their factories and supplY chains, ~ i t
special ordering, a customer would go to a dealer and specifv
car with a particular set of options. The car would be built on
the arder was placed. Custom ordering a new car was once a
annual or biannual ritual for many Americans and canad
¡S now much less common. As the retiring general manager of
sales division at a Big Three cornpany recently told us with so
satisfaction, "lf I've accornplished nothing else in my years he
I have succeeded in stamping out special orders!"
~h~ European and Japanese exporters to the United States
never accepted special orders, because of the distantes ind in supply. Rather, they concentrated on adding a variet.v
equipment o11 the cars t h e ~export. With
gro,Yth of irnport franchiscs over the years, consumers have
y more choices. In 1958, for example, U.S. consumers couid
twenty-one different makes of car from ten different manufaculd buy 167 different models sold ~lnder
&es from twenty-five difkrent producers.
n consumer now has an enormous and
ts to buy off the dealer's lot. Horvcrier., if
vailable that fits the customer's required it difficult to special order ene.
European distribution system has, in manv XvaYs, closely
mbled the American, except in many respects it's thirty F a r s
Europe, there are not only more dealers than
the united ~ t a t e sbut
, in many countries there is still a two-ti~r
aler structure, something that disappeared from the United
. That is, in addition to Europe's 36,200 mail1
Iso 42,500 ~ubdealers.~
Most of these are small
new cars that are supplied through the *sin
saler. Compared with the United States,
eIIs, on average, 393 cars a year, the average
ern Europe sells only about 280 cars a year,
, if we count the subdealers, only 128 cars a year e x h . (DaVd
dealership in the United Siates .rvith Europe
apan are given iil Figure 7.2.)
ope's system also has the additional complication of another
between the manufacturer and the dealer, namely,
pany, which performs many of the funcas overseeing dealers-of the regional sales office in
united States. However, in many cases these coinpanies are
owned by the manufacturer. For example, Volvo cars are sold
tlle U.K. through Volvo Concessionaires, which is o ~ ~ n I->Y
e dthe
that also owns many car dcalers in the
performance sports cars. As par1 of a deliberate strategy of
ing
differentiatc these cars from those offered by the vol
~roducers,some specialist producers instituted a much hi
leve1 ofservice lo the customer. For example, Volvo, together
its U.K. importer, Volvo Concessionaires (owned by Lex servic
piotleered a lifctime service contract and other forms of enhan
care.
Tliis approach was rapidly copied by other European spe
ists and has becn applied in their dealerships in ~
~
~
t
For examplc, part of the s~icccssfulsales recovery of ~a~~~~
North America in the earlv 1980s was through enhanced custo
carel which overcame consumer concerns about the reliabilit
the product. Recently the new Japanese luxury brandsLexus, and Infiniti-have pushed farther down this path
quiring that their dealers spend large sums on dedicated sel
sites built to standard designs and on staff training,
European specialist producers have also continued and
encouraged custom ordering of cars. In the German dom
market, for example, Mercedes offers no option packages; ra
al] options are "free standing" and installed at the facto
customer order. (The inability of the factory to do this eas
accurately is, in fact, one of the causes of low productivity in
European specialist plants we visited in Chapter 4.)
However, in other respects the basic dealer stmcture
remained the same, even in these luxury dealerships. ~t is ta
as a
that a high leve1 of service entails high distribut
costs and that this can only be justified on luxury makes wi
high gross margins. Cheaper cars can logically only be so
through dealers offering the minimum of assistance to the co
sumer.
l
1s there an alternative, lean approach to selling and servici
cars, an approach that completes the lean-production system?
we think there is, at least in logic, and a number of
elements can be seen in Japan toda).. The Japanese system is
an ideal model of a lean distribution system for a number o
reasons and, in fact, as we'll see shortly, it is changing. Howeve
the way the Japanese producers think about distribution in the
estic market and the manner in whicli the pieces of their
em [it together point toward tlie lean distribution system of
h
future, a system hardly imagined in the West.
T~ see lean distribution for what it should he, we can't begin
a rarrow cost-cutting perspective, the normal Western ap~ h in, which factors such as the ilumber of sales made by
salesperson per month are the way to measure success.
her, we must view i t a s an essential component of the entire
roduction system.
et's begin by visiting a typical car dealer in almost any
ern country. The premises consist, effectively, of a large
ing lot on which sits a vast array of new cars gathering grime
rnnning up interest costs. The sales personnel. who work on
commissions, live on a fixed proportion of every sale
a small base wage. Most are prolessional sellers, not product
&lists. ~ h is, ~they've
t
received their training in sales techues, particularly in how to drive an effective bargain, rather
in the special features of what they're selling. So it really
't matter if they're selling shoes, computers, encyclopediac,
we've visited showrooms for years as part of our IMVP work
d are continually amazed at just how little salespeople do know
out their products: The salesman who defended the merits of
r.wheel drive in the front-wheel-drive car he showed us; the
eswoman who argued for the fue1 economy of the four-cylinder
ine in the V6 model on display; the salesman who volunteered
shoes, his previous product specialty, were a lot easier to se11
n rhe cars he had been selling for the past two weeks-these
only a few exampies of saiespeople's glaring lack of product
wledge, a problem that is particularly acute in North America.
e turnover of sales personnel in Europe is much lower, and
y seem to know more about the specific products they are
lling.)
While it mav still be possible to special order some makes of
in North America, the sales staff pushes the CuStOmer ves.
rd to take a car already on the lot, perhaps by offering a better
scount. once a deal is struck, after some intense haggling, the
stomer, now the buyer, is turned over to the financia1 staff to
rige payments, then to the service staff to arrange delivery.
service department is in charge of taking care of any subse-
througb the channel are being developed, staff members from
questionnaire from f h e assembler. "Were you satisfied with
ehannel
are on loan to development teams. These channel
car and with the dealer?" the company wants to knoiar. ~ ~ d ,
esenlatives
are in a position to malte an invaluable contribusome years alter the sale, the buyer is likely to receive a mont
to
product
development, los reasons we'll examine in a
or quarterly magazine from the assembler with a few gene
interesl asticles and information on new products.
he charinel, nihich is part of the Toyota company, sells its
That's the extent of tlle relationship between buyer and
hrough seventy-eight dealer firms, each with aboui sevenfor the most expensive consumer purchase most ,[us make in
ifferent sites. (This contrasts viitli the several hundred or
lives. (Remember that our homes, the other big-ticket item in
Personal consumption, usually appreciate in value,
our
rcent of tlle dealers are o\vned by the Corolla ~ h a n n e l .The
'~
depreciate over a decade or less to near worthlessncss, so
are
either
partly
owned
by
Corolla
or
independentl~
owned,
teriTis of net consumption, cars are much more important th
the training is done centrally by Corolla. Each of
housing.)
ough
e dealerships has had a long and close relaiionship with
How does the seenario we just sketched contrast with t
,,ta,
and they can best be characterizeci as par1 of an extended
sales practices of the lean Japanese producers? Again,
la
. y that is fully integrated with the parenf. In addirion
Toyota as an e ~ a m p l eToyota
.~
has five distrjbutjon "channels'
.ng, thc ehannel provides staff and a full range of services for
Ja~an-Toyota, Toyopet. Auto, Vista, and Corolla, and is abou
dealerships where i t does not own the facilities. ,411 told, the
opcn a sixth. (Nissan and Mazda also have five separate chann
nne]
so]d about 635,000 cars and trucks in 1989 and has 30,400
arid Honda and Mitsubishi have three each.) ~h~ ,-hanne
simply the name on the dealership. In the United States,
~h~ elnp]oyces, many ol whom are college grad~lates,are
example, we might see the name of the individual 0wner-l
ed ,.ight after gradualion each spring. They undergo an intenSmith Buick, say. In Japan, it's Toyota Vista os Toyota coro
e training program a t the Corolla "University," which offers
The channels are nationwide and in many cases are o\vned by
mostly related to marketing. Once the new employassembler. Each channel sells a portion of the total T
O
~
~
~ ty courses,
~
are
fully
trained-although
formal training continues every
uct range. For example, one channel may se11 less expen
for
every
employee-they're
assigned to specitic dealerships
models, another sportier ones, and so forth.
begin
selling
cars.
The channels have different labels and model names for th
yhe sales staff in each dealership is organized into teams of
cars. but the main thing that differentiates them is their app
eight, an organization ve- similar, in fact, to the work
en
different groups of customers. Since every car in al1 tive ch
in the Toyota and NUMMI asscmbly plants we described in
nels is clearly identified as a Toyota, the purpose the chan
ter 4. just like those in the Sactory, these teams are multiis noc to establish brand identity, as with sales divisions in
led; al1 members are trained in al1 aspects of sales-product
United States. Rather, it's to develop a dircct link belween t
formation,
arder taking, financing, insurance, and data collecmanufacturing system and the customer-whom Toyota, not a
which we'll explain shortly). They're also trained to systemcidentally, calls the owner.
ly solve o~vners'problems as they arise.
To see jusL how the Toyota system works Jet's looli at one
work team begins and ends the &ay with a tealn meetthe five channels, Corolla. Toyota established this channcl in 19
During the bulk of the day, team inembers disperse to
to se11 the Publica model but changed the name to coro1la in
door.to-door, with the exception of one team that staifs the
when this new model replaced the Publica in the ~ o y o pro
t~
ormation desk in the dealership. Each month, the entire team
range. It has since expanded its lineup to includc the sup
a day to solve systcmatically any problems that ha17e
C a m ~ Celica,
,
and Corolla II models and thc TownalI van a
ed up, using the "five why's" and other problem-solving
pickup.
niques. These meetings are the sales equivalent of the quality
The Corolla channcl is directly tied into the product.deve
rcle in the Sactory.
ment ~rocess.During the entire period that new cars destined
Selling of cars door-to-door is unique to japanand is
f o r n l l ~bewildering to foreign observers. H~~~~~how it
Team members draw up a profile on every household wi
geographic ares around the dealership, then periodically
each one, arter first calling to make an appointment, ~
)
visits the sales representative updates the household profile:
manY cars of what age does each family have? what is the
specifcations? How much parking space is available?
many children in the household and
use does the f
of its cars? When does the family think it will nee
replace its cars? The last response is particularly importa
lhe product-planning process; team members systematicall
this inforination back to the development teams,
On the basis of the inlormation they've garnered and a k
edge of the Corolla product range, the sales representative
gests the most appropriate specification Sor a new vehiele to
this particular customer's needs. The family may have do
about ~ h a tot buy, ofcourse, even ifit's really in the market
ea-, so the sales representative may bring a demonstration ve
the next visit. Once a household is ready to buy, it plac
special arder through the sales representative. A vast majori
cars in Japan are customer ordered, just as this possibilit
being eliminated ir1 tlie United States. ~h~ vehicle
l~Picallyincludes a complete financing package, trade-in
car, and insurance, because the sales agent is trained
provide one-stop service Sor the auto buyer,
If cars are customer ordered, you might a&, how does
factory cope? Nere's what happens.
Factory executives try to make an educated guess
demand for different versions, colors, and so forth, on th
of this forecast, they establish the plant's build schedule,
I h e ~also give to the components suppliers so the latter
u'hat to make. The accuracy of these forecasts obviously depe
how frequently the build schedule is revised, ~ h is itypica
~
everY ten days in Japan. compared with every month to six we
in the West." Once the 01-deis come in, the assembler adjusts
build schedule to make the specific cars the customer wan
Because the Japanese practice just-in-time production, do¡
is much easier than in the West, which has much less fle
factories and much longer lead times for ordering parts (which
around as inventory for a long time before they are used).
Of course, the Japanese build schedule is more accurate
~
rst place and can accommodate a customer-specihed arder
easily because of the much ckuicker feedback fron1 [he
mers about what they re&Iy want, and because Japa*ese
watch on trends in tastes. Tlie asseme keep a much
the components
suppliers can plan &ead more
~ lanti and ~
~
the right mix of products &oing down
rately anct
-for example, mixing some high-~~ecification
cars that take
tle longer to build with low-~~ccificaiion
cars that ?alce
in
e, Japanece factories can cielivcr a customer-ordcred
two weeks in japan.The same order-if the cUstonler could
all-would take six weeks in [he WesL, at besi, and couid
long as tl-iree months.
at about determining a price? Because the customer is
a car lailored to bis or her needs, ihe haggling h i t
ar buyers find so distasteful is aimost eliminated il1 rhe
system, ~h~ salesperson doesii't need to discourli ihe
in arder to get rid a car that the customer w(luld rarher
. Moreover,
prime objectiie of Japanese dealers Lo
e customer feeling that he or she is part os the dealerPs
~
) want ~custoiners
~ to lthink that
~ Ihey
~ have~been
d well and have paid a fair pricc.
member that this is likely to be one of many transactions
customer does \"ith this salesperson. The sales~ersonwill
have sold-the customer a car in thc past, laken
of the formalities o[ registering it and disposing of the traded
arranged to have the car serviced, and seen i t thr"ugh the
ous government inspections. Quite possibly the salesPcrson
lso have battled with tlie insurance company for an accident
on the customer's behalf and lent him or her a car while
customer's own car was being repaired. In the West thc
on to make the most out of a one-off transactio*
sure
commiteen two strangers with no subsequent loyalty
( E if ~
the customer
~ ~
returns to the same dealer sor a
purchase, the salesperson is Iikely to have moved on.)
he J~~~~~~~system the aim is to maximize the stream of
from a customer over ihe 10% term.
with very few defect~in Japanese cars and intense comPetiin the japanese
car market, it is clearly understood that the
le, will fix any problems the owner encouniers niith the car
after the end the formal warranty. The customer need
e with dealers to get them to accept responsibility for ~ ~ a r y claims, an unpleasant experience that usually con\'inces
Western customers to look elsewhere for their next car,
larly to a brand that has a reputation for few gaws. onc
contract is signed, the order goes directly to the factory. when
car is r e a d ~ in
, ten days to two weeks, the sales representa
~ersonallydelivcrs it to ¡he new owner's house. The new car bu
need never go near a dealership.
The lean Dealershi~
Some Japanese buyers, particulariy younger ones in big
would rather visit a dealer. Unlike older people, theyjre m
interested in shopping around and seeing for themselves
~ r o d u c t sare being offered. This trend is occurring just as ma
facturers are finding it increasingly difficult to recruit
willing to se11 cars door-to-door. For one, the assemblers
hiring more women as salespersons, some of whom are not ea
10 knock on doors. particularly during the evening. ~h~
More and more Japanese are buying their cars at the dealershj
about 20 Percent do so in the Corolla channel, and this percenta
is higher for other channels. In addition, as we'll see, all
xTisilthe dealer eventually to get their car serviced,
A typical modern Corolla dealership may look si
Western dealerships in one 1-espect-its sl1owroom-b
thing else is different. For starters, no vast parking areas
fact, you'll see few cars on the premises other than thre
demonstrator models. Since most cars are manufactured
there is no expanse of finished vehicles to buy off the lot and
sixty- or seventy-day stock of cars running up interest costs,
Japan, the stock of finished cars in the system averages o
twenty-one days.12
Second, a sales~ersondoesn't descend on the hapless shop
per. Since the team is paid on a group commission, the seven
eight team members in the showroom have no incentive to gr
the customer before the next salesperson does or suggest that
Orshe can provide a better deal. Instead, members of the tea
Join in the discussion once the customer approaches them to a
a speciíic question.
The heart of any Japanese dealership is its service asea.
asea's rimar^ purpose isn't to rectify problems
cany
routine servicing as is the case in Western dealerships, ~ ~ t h
pare vehicles for the Ministry of Transport
t provides a major source of rei~enue.(Govent inspections also exist in a milder form in Europe and are
mild. by comparison with Japan, in North America.) Al1 c a n
pass the first inspection when they are three years ola. The
stry then requises inspections every two years, unti1 a car's
inspections are requircd annua1l.y.
spections becomes quite high as cars ase.
frequent, but more ciemanding. For
th year the entire brake system will
t, even if it functions normalb. So the
ncentive to buy a new car after four ~ e a r s ,
most japanese
retire their cars at this time. (The dealer reselis
a third of the trade-ins in the local market. Another third is
east Asian countries for sale there, and
ecause the repair costs the dealer would
As
ur to fix them to meet test standards are prohibitively
'11 see, apart from wear and tear, dealers fix anything that @es
ong, so they don8twant to take the rislc of taking on high fu1u1-e
sts tarnishing their reputation.) Car buyers have little oPP'Jrnity to postpone their next purchase in hard times-a common
sponse to economic downturns in the West.
Ghannel Loyaliy in Lean Production
that buyer loyaity to a given brand is
ely a Vestige of the past in the Western auto markets. The fact
a customer buys a ChevroIet or a Renault once doesn't
case the likelihood of that customer's buying a Chevrolet 0s
round. Far from it. Most Western consumooking for a good bargain or an available
the u.K., for example, brand loyalty has fallen from around
to 50 percent today. It's even lower in the
ese, moreover, repurchase of the same brand
umer's age-from about 30 percent for those
rcent to 23 percent from ages twenty-six
percent los those under age t ~ e n t y - f i v e . ' ~
oesn't exist in Japan. A key objective of everY
distribution channel is to build and nurture lifetime chann
loyalty. Again, let's look a t the Corolla channel.
Once a new car is delivered, the owner becomes part of t
Corolla family. This means Srequent calls from the person selli
the car-who henceforth becomes the owner's personal sa
agent. The representative will make sure thc car is wor
properly and ferret out any problems the owner may be having
relay hack to the factory.
The sales agent also sends the owner a birthday card
condolence card in case of death in the Samily and will cal1 t
if sons and daughters will need a car as they leave for college
their first jobs. It's often said in Japan that the only way to esca
the sales agent from whom you once bought a car is to ieave t
country.
One aspect of this relationship would no doubt be parti
larly welcome to Western car buyers. Because the channe
obsessed with market share and tries never to lose a single owne
the relatively short-term warranties Japanese assemblers offer a
ignored. The channel will, in general, continue to fix defecti
cars at no cost to the owner throughout the vehicle's normal Si
provided the owners have not abused them. (Obviouslv. this i
plicit warranty doesn't apply to routine wear, such as replacin
brake and clutch linings.)
LEAN VERSUS MASS OISTRIBUTION: A SUMMARY
AS we'vc seen, the lean approach to dealing with customers
significantly different in concept lrom the mass-producers' a
proach. First, the Japanese selling system is active, not pas
indeed the Japanese cal1 it "aggressive selling." Rather
waiting a t the dealership for customers attracted by advertisi
and publicly announced price cuts, such as factory rebates, t
dealer's personnel periodically visit al1 the households in t
dealer's service area. When sales lag, the sales force puts in mo
hours, and when sales lag LO the point that the factory no Iong
has enough orders to sustain full output, production personn
can be transferred into the sales system. (This type of trans
occurred during Mazda's crisis in 1974 and, more recently. a
Subaru.)
Second, the lean producer treats the buyer--or owner-as a
ntegral part of the production process. The elaborate data collecon owner preferentes Sor new vehicles is Sed systematicall~
development teams Sor new products, and thc c o m p a n ~goes lo
traordinary lengths never to lose an owner once he or s h is
~ in
~ h i ~ the
d , system is lean. The whole distribution system
tains three weeks' supply of finished units, most of which are
he system that delivers this high leve1 of service is also IrerY
fferent from a mass-production dealer system. Thc industry is
y ~ n u c hmore concentrated-there are only a total of 1,621
aler firms in Japan, compared with some 16,300 dealer princials in the United States, a market two and a half times larger
an ~ a p a n~. l m o sal1
t Japanese dealers have multiplc outlets and
me of the largest easily match the megadealers Sound in the
ited ~ t a t e s .In the same way as lean producers only have a
ited number of suppliers, they only work with a limited numr ~f dealers, who al1 form an iniegrated part of their leanoduction system.'"
THE FUTURE O f LEAN CUSTOMER RELATlONS
f many elements of the Japanese systcm are superior, as we think
hey are, why haven't they been copied in the West? When we
sked the Japanese lean producers and the Western mass-produrs this question, two completely different answers emerged. The
stern producers unanimousl>r argue that the system is too
pensive, "a cost-control nightmare they wish they could get out
,,
in Japan, to quote one. It requires large amounts of effort to
11 each car, they maintain, as shown by the fact that the average
les representative in a U.S. dealership sells ten cars per month
r about one car every two days), while the average Japanese
les representative sells four cars per month (or about one a
ek). From the mass-producer's perspective, in which tlie costs
selling are already too high, this extra cost seems completely
possible to justify.
The Japanese perspective is quite different. First the systein
f door-to-door selling is viewed as an anachronism suited to
pecial conditions in Japan. It is gradually being phased out there,
o it is only the other elements of the system that thc Japanese
would wish to introduce in North America and Europe. Howeve
and most important, as one Japanese executive pointed out: "T
systcn~makes no sense unless cars are built to order and deliver
alnlost immediately. We can do this only as \ve develop a co
top-to-hottom manufacturing system in North America an
ropc bv ihe end of the 1990s."
In the current situation, where Japanese cars are manuf
where Japanese companies are constrained by market-share 1.
itations in rnany markets, they've chosen to behave like west
mass-producers. We believe that following the Western syst
not their ultiniate intent, and that the Western mass.prod
may be in for a final surprise before the end of the 1990s, as t
las1 elemeslt of the lean-production system emerges,
The Japanesc companies are quite aware of the costs of th
system-no one is better a t analyzing the costs of evev step
production down to the last yen. They argue that these tos
\vould make 1-10 sense if lean selling were serving the same fun
tions as mass-production selling. However, they point out, it do
much more. The lean selling system, with its periodic surveys
practically ali consumers in the Japanese market, is the first ste
in tile product-development system. It avoids the need for
time-consuming, expensive, and frequently inaccurate mar
assessment surveys of the Western mass-producers.
The lean selling system also reduces inventory costs drama
ically and smooths the flow of production in the [actov.
making sure its sales force has a clear understanding of the nee
of the factory, in particular for a smooth flow of total orders
. . snix os orders fluctuates, it's possible to make the fact
as the
work better.
Moreover, the Japanese system helps fine-tune new pro&
and
embarrassing or dangerous errors before m a ~ s i v e - ~
highly visible-public recalls are needed.
Finalb, the lean selling system instills channel loyalty in t
buyer and makes it extraordinarily hard for new competitors
gain share. This is a key reason Western m a ~ s - ~ r o d u chave
er~h
such a hard time making headway in the Japanese market. ~t
only in the past few years, as ~ e s t e r nfirms such as BMW
Daimler-Benz have made the necessary iilvestment in their
distribution channels, that import sales have become
rising 5 percent in 1990 compared with less than 1 percent
decades.'j
we've pointed out, the Japanese companies are well alvare of
costs, particularly for door-to-door sales, iust as thev
ware "f their costs in every other area of production. Thev
e that the most promising way to cut back these costs lies in
informatioll lechnology. To see how informalion tech.
might work, let's take one more trip to our Corolla dealer.
e first tliing a consumer encounters on entering a Corolla.
rship today is an elaborate computer display. Each Corolla
ner in tlle ~ o r o l l afamily has a membership card that can be
erted in the display, just as one would insert a card in a
&ine. ~h~ display then shows al1 the system's information on
at buyerrs household and asks if anything has chailged. If it has,
e machine invites the owner to enter new information. The
stem then makes a suggestion about the models most a p ~ r o p r i e to the household's needs, including current prices. A sanlple
model is usually on display in the showroom immediatelv
jacent to the computer display.
~t this point, if the owner is seriously interested in buying, he
she can approach the sales desk where the seven or eight team
and discuss the particulars os a sale. Cars
embers are
20
~d in this maslner are rising steadily in Japan (from
rcent at present), and the companies hope that over the lon&
m, they can largely deal with most existing owners in this way.
deed, at some point in the future, they hope that the same
rmation will be available at every owner's heme on a comter or television screen.
~h~ customer has access to other data bases as \vell-on
rything from obtaining financing and insurance to parking
ermits (necessary in many Japanese cities before YOU can buy a
1. customers can also access inlormation on secondhand cars
uld they want to buy one and obtain data on al1 their trdnsacwith the dealer lor service and inspections.
owner wilI still know an individual in the sales
while
twork to contact in case of difficulty, most of the sales force can
be directed to "conquest" sales of owners who are currently
yal to other brands. The end result, the companies hope, should
that the selling cost for the average new car will drop substanlly, but the information harvested from consumers and the
feeling of channel loyalty will be retained. If the Japanese pro
cers can accomplish this goal and then transfer this truly
selling system across the world, the system of lean produ
will be complete.
What we see in Japan, then, is that distribution is a
integrated part of the entire production system. It is not si
an expensive door-to-door selling system. In essence it is a sys
that provides a high leve1 of service to the customer and a
leve1 of real feedback to the manufacturer. When product
ning, marketing and distribution costs, together with th
of more accurate matching of production to demand
less discounting and distress selling) and better production sc
uling (meaning a more efficient factory) are added in, the Ja
nese system is already delivering a higher leve1 of service
much lower real cost than Western analysts have realized. W
information technology is fully added to the system to produ
truly lean distribution network, it should be possible to efimin
yet another of the inherent compromises in mass production.
as quality costs less, not more, in the lean factory and as desig
products faster reduces costs and errors, selling cars in a l
fashion with a high level of service should be possible at mu
lower real costs than in mass production with its low leve1
service. Lean distribution will form the front end of a system t
is driven by the needs of the customer, not by the needs of
factory. In an increasingly competitive world market where
we saw in Chapter 5, more affluent customers are seekingahle to pay for-a greater choice in personal transportation,
reorientation of the entire mass-production system will be crit'
for survival.
Today there is much discussion in the West of the inadeq
cies of the distribution system. Customers are unhappy, the
semblers are unhappy, and the dealers are only marginally p
itable. However, discussion in the West of the future of autom
distribution has so far focused on finding a new winning fo
for dealers-megadealers, publicly owned dealer chains, sepa
sales and service outlets, or lifetime-care schemes that kee
customer coming back for repairs for more than the first
years of the product's life. As we've seen, however, this isn't
right way of looking at the issue. Instead, we must thin
distribution in a wider context, as an integral part of a custo
focused lean-production system. The winning formats that fi
system may turn out to be quite different from our cu
ectations. lndeed, we may end up with not just one winning
at but several to suit different types of customers, products,
market segments.
with this revie\.v distribution, we've now looked al al1 the
in the immensely complex task of producing a motor "ehione of the features of lean production we noted in eveT
pter was the need for elose coordination between the manY
often involving face-to-face contact. SurprisinglY, this is
of distribution, where a iruly lean distribution sYstem
requires a production system in or very near the market
B~~~~~~"f the great distantes between the world's major
rkets and the persistente of trade barriers, this suggests in
that lean producers wishing to succeed in the global motor
the long term will need to develop complete Producstry
,distribution systems in each majar region. But how does a
mpany create and manage such a global network of production
l-hiS is the challenge we'll examine in the next chap-
bank were the major source of funds for investments by t
companies in the group.
The Americans eliminated these tightly organized groupi
f
during their post-World War 11 occupation of Japan. ~
Americans left, the zaibatsu were replaced by a new
industrial finance, the keiretsu. Each Ieiretsu consists of
twenty majar companies, one in each industrial sector.
zaibatsu, there is no holding company at the top of the orga
tion. Nor are the companies legally united. Rather, they're
together by cross-locking equity structures-each company
a portion of every other company's equity in a circular patte
and a sense of reciprocal obligation. Toyota, for example, is a
ated with the Mitsui keiretsu, while Mazda is a member
tomo, and Mitsubishi Motor Company is a member of ~i
Among the key companies in every group are a bank, an insura
ComPany, and a trading company. Each of these has substan
cash resources that can be made available to the members of
group. In fact, their key purpose is to help each other
investment funds.
These groupings arose only gradually as Japan rebuilt a
the Americans ieft. The equity in the previous zaibatsu had b
dedared void in 1945. The Japanese companies initially
financed almost entirely by loans supplied by the big 'rokyo
and guaranteed by the American government. Since com
had only these loans and their physical assets, their equity
very modest. As the economy took off and many companies
carne profitable, they began to wony about being bought up
foreigners. They also distrusted the arm's-km& stock mark
the primary means of generating equity, because they coul
imagine a system in which there was no reciprocal obligation
To address these concerns, the growing companies of
1950s and 1960s hit upon the idea of selling equity to each oth
often with no cash changing hands. So each member of the prew
grouPsr and some newcomers as well, joined the new keivets
which the equity went around in a circle.
The large groups were essentially privately held-but o
massive scale. That is, their stocks traded in small volumes on
highly volatile Tokyo stock market, but the stock that
counted was never for sale. The Americans and other foreign
discovered this fact after 1971. That year equity in J~~~~
liberalized to permit foreign majority control of any compan
t
~
t nene of the keiretsu members was willing to se11 its "captive"
res any price. SO few companies actually could be bought.
The system was glued together partly by a seme of reciprocal
~ ation-ach
member of the group held every other member's
in a sort ~f trust. However, if the sense of obligation faltered,
more practical factor of hostage equity kept selling in check:
e company considered selling its stake in another to an
er seeking to gain control, the second company could retaly selling the first company's equity to outsiders as well. No
A variant of this system was soon extended downward into
supplier groups, as well. We saw in Chapter 3 how Toyota sPun
supplier companies, such as Nippondenso and Toyod'd Gosei.
yota held an equity stake in these companies, and they had a
11 equity stake in Toyota. Soon the Toyota industrial group
bited some of the circular equity structure seen in the Iteireisu,
ough Toyota held a strong position at the center.
~h~ recent attempts of the American raider T. Boone Pickens
seize control of Koito, a Toyota group member, show just how
werful the group system is. Toyota owns only about 15 Percent
Koito, and Pickens was able to buy up shares totaling more
an 26 percent. ~ e the
, couldn't gain a seat on the Koito board.
&dition, no other shares seemed to be for sale, even at an
ring price well above what the shares would fetch on the open
~ hsystem
i ~ of group equity has been exasperating to Western
mpanies and governments because its logic is so different.
se companies, with what at first appears to be a public
stmcture, are in reality privately held. This arrdngement
ould not be allowed under the investment laws of the United
ates and a number of European countries-companies would
e to explain that only some of their stock was actually for sale.
.le we believe the keiretsu and industrial groups are in fact the
t dynamie and efficient system of industrial finance yet deed, they are not adequately understood in the West.
id^^ providing members protection against hostile takecited advantage of the keiretsu system is the low
vers, an
of funds for group members. The inexpensive funds come in
forms.' First, many Japanese companies pay hardly anY
idends. Typically, they pay a 10-percent yield on the parvalue
their stock, where tbe par value, established at the time of the
MANAGING THE LE4N ENTERPRISE
initiaJ equity trades in the 1950s, is substantiafly zero. soT~~~
stock, for example, in fiscal 1989 paid a dividend of 18.5 yen or 1
percent of earnings, while Nissan paid only 7 yen or 7 percent
earnings.
Second, in the 1980s, the booming Tokyo stock market pe
mitted thc Japanese auto companies to issue vas1 amounts of
stock in the form of convertible bonds that could be converte
stock if a company's stock reached a certain price in the ma
Buyers of these bonds \vere ivilling therefore to accept
interest rates on the presumption that their real re1ui-n wo
come from the stock conversion in the continualfy
Tokyo market. During the 1980s Toyota issued $6.2 billion
convertible bonds a t interest rates from 1.2 to 4 percent, a cost
capital far below that available to Western auto companies,
much weaker companies such as Isuzu and Fuji Heavy xndust
(Subaru) were able to obtain low-cost financing through
means.'
How long this second form of low-cost fund raising can
t i m e is an interesting question. On the one hand, the
the T o b o market in 1990 suddenly made investors aware th
conversion will not always be possible, and the issuance of co
vertible bonds stopped, at least temporarily. ~n the 0 t h han
J w a n is still a country of obsessive savers, and these savings mu
find some outlet.
However, even in the absence of cheap investment funds,
Japanese groUP systeln stiii confers a significant competitive
SPent. For proof of this hypothesis, we need look no farther tha
the waY financia1 systems in Japan and the West deal with com
panies in distress. The key Japanese instance is the Mazda turn
around in 1974. Up to that time, Mazda had heen run by .
founding family, which was strongly oriented to product en
neering. The hallmark of the company was its fuel.hungry
technically advanced Wankel rotary engine. When energy pr
suddenly soared in 1973, Mazda faced a major problcm. 11 neede
an entirely new set of fuel-eflicient piston engines, and it neede
a new range of models for the new engines.
The c o m P a v faced anothel- pressing problem as well. Maz
had been charging higher-than-average prices for cars in its m
ket segment, even though the Wankel was cheaper to makt: th
ordina1-y engines. The reason for the high prices was ~~~d
inefficient production system, which resembled mass production
197
more than lean. Up to this point, Mazda's cars could
mmand premium prices because of the high-tech image of the
motor. scrapping the Wankel meant Mazda would llo\v be
&ing ordinary cars, so prices had to come down. And for this
rop to occur, it was essential that Mazda reform its productiori
Mazda8s salvation came from the Sumitomo group, ~ h i c h
ntrolled t1ie car compariy's equity through cross links. The
irnitomo bank sent a team of executives to replace the famili
anagement. These new executives' lcey decision was to col?! thc
yota production System at Mazda's Hiroshima production comex, so that Mazda could become cost- and q ~ a l i t y - c ~ m p e t i t i ~ ~ e
ith the best Fapanese companies. A second I<ey decision \vas to
rovide massive koans for new engines and a new model range, so
~ ~ uexpand,
l d
rather than contract, its market presencc.
~h~ contrast with British and American practicc in the 1970s
d 1980s is striking. When British Leyland and Chr~slerbcgan
founder, their bankers and institutional investors, numbering
the hundreds, were mainly concerned with how to minimizc
heir exposure. In hoth cases, tlie stock was wideli held and no
effective organization of stockliolders existed to voicc their conems. The outside members of the company boards neithcr undertood the real problems nor knew what to do. fnstead, the boards
rcmained passive, the banks wrote off their loans, and the instiutional investors simply sold their stock a t a loss and walked
Brjtish Leyland ended up under direct government control
for a decade, while Chrysler needed a government-guaranteed
lean to regain its momentum. Hocveirer, neithcr firm inspired
enough confidente in governments or investors to quaiify for more
than the minimal amount of assistance to keep going. With
constlaints on product-development funding, both companics
struggled through the 1980s. More significant, in neither case was
the hnanciaI system or the goverriment able to tackle the real
oblem: dysfunctional mass-production systems that could no
nger compete in the world market.
The investment finance systems behind the other Western
producers in Europe have been more effective, aL least in providing companies the funds necessary to weatlier crises. This is
because a single large shareholder with a long-term outlook has
controlled the destiny of most European companies: the Agilelii
family at Fiat, the Peugeot and Michelin families at PSA, the
Quandt famifyat BMW, the Handelsbank a t Volvo, the wallenber
Fdmily at Saab, the PorscheiPeich family at porscbe, and
Deutsche Bank at Mercedes. Renault, of course, is still
~ w n e d and
,
Volkswagen had a lai-ge government shareholdi
~ n t i recently.
l
Thus no company has heen "friendless," with
significant shareholder committed to ttie company and no stron
relation Lo a major bank.
Nevertheless. although the Japanese groups do make m
takes, in some instantes very big ones, the keiretsM system
average has exhibited superior performance compared with b
the Anglo-Saxon (American and British) 2nd continental
Pean SYstems of financc. Western finance ten& to be &her imp
tient and largely uninformed about a ~ o m p a n yproblems
'~
(as
the case of the American and British institutional investors an
banks that dump their stock and loans at the first sign of troub
~ a t i e n but
t passive (as in the case of outside directors in t
United States and Britain and the f'amily shareholders in con
nental Europe). The latter have often failed to confront the pro
lem of clear slippage in competitive position unti1 very late in ih
game.
BY contrast, the Japanese group system is patient and
tremely long-term in orientation-but v e r - well informed
h i g h l ~critica1 of inadequate performance. The groups can a
t" invest heavily tu finance corporate turnarounds, because
considerable knowledge reduces the risks of failure.
CAREER LAODERS
As we've noted at a number of points, mass production provides
no career Progression for production workers. Engineers, financial
a n a i ~ s t s ,alld marketing specialists progress through technical
exPertise. The Progression for the general manager is through
1
paths
higher levels of the corporate hierarchy, ~ 1 three
are dysfunctional for the organization as a
~h~ lean enterPri% by contrast, strives to provide every employee with a
career path, although these are very different from those of mass
production.
T" begin with, every employee begins by working on tiie
Production line for sume period of time. F~~ example, while
"isiting the Honda plant in Marysville, Ohio, recently, we asked
meet with the externa1 affairs director, the person at Honda
handfes relations with governments and the general ~ u b l i c .
as unavailable, we were told: He hha just joined the ComPanY
was busy assembling cars. The best lean producers believe
production is where value is truly added, not
at the point
rough indirect managerial activities, and that al1 employees
eed to understand this fact as soon as they enter the comPanY.
~h~~~who stq in the factory grow iricreasingly able to sol\'e
blems. Management siresses that problem-solving is the most
job. Management's objective is to give
ortant part of
ployees increasingly challenging problems to solve in arder lo
continually their skills, even when, unlike in Western ComPano promotion up a ladder to section head, say, or factory
nager is possible. Higher pay comes largely on the basis oC
bonuses as well. In other words, the
rity, w i ~ h
manufacturers, which operate without much of the hierarchy
find in ~ e s t e r n
companies, try to make employees understand
t their capaciiy to solve increasingly difficult problems is the
meaningful type of advancement they can achieve, even if
ir titles don't change.
F~~ those employees \vith a specialized skill-mechanical
gineering is the most common-the lean producer attenlpts to
rness the skil] to a team process where it will be of maximum
e, we saw how this technique works in Chapter 5 . We also saw
team members are shifted to subsequent tcams and how they
be asked to 1eal-n entirely new skills as they move through
F~~ employees who are needed for general management, the
ntrast between mass production and lean is equally striking.
se decision-making and problem-solving are pushed far
[he ladder in the lean company, it has much less need
and senior managers to send orders down lhe
ers of
rarchy and transfer information back up. Instead, the k e ~
nctions for managers are to tie the supplier organizations to the
organization and to tie together geographically dissed units of the company. Typically, the company sends managers at the rnidcareer leve1 to high-Ievel positions in the supplier
companies in the assembler's group and rotates mid- and seniorlevel managers between the company's operations, particularly
the foreign operations.
l-hese practices have two advantages. They create a complex
network of interpersonal relations, so the assembler and the
s u ~ ~ l i e and
r s those in the company's international operati
know each other through personal contacts. They a r
conduit lhrough which the company's culture is spread into t
supplier system and to new regions.
6EoGRdPHlC SPREilD
The world at large, including, we must note, a number o
Japanese lean producers, does not yet understand a vital fea
of lean production. This mode of production achieves its hig
efficiency, quality, and flexibility when al1 activities from de
to assembly occur in the same place. As a senior Honda execut
recently remarked, "We wish we could design, engineer, fabricat
and assemble the entire car in one large room, so that
involved could be in face-to-face contact with everyone e
as we saw in the last chapter, the final step in the system,
sales and service, cannot work a t al1 without a production sys
located in the same area as the sales market,
For this reason, lean producers in the 1990s will need
create top-to-bottom, paper-concept to finished-car manufactu
ing systems in the three great markets of the world-North
ica, Europe, and East Asia (centered on Japan). This process
farthest along in North America, where the Japanese firms beg
o ~ e n i n gassembly plants in 1982. Eleven were in operation by t
end of the decade and in 1990 accounted for slightly more th
Percent of ~ u ~ o m o b iassemblies
le
in North America, as sh
figures 8.1 and 8.2.
Doing the whole job in one large room wilf not be Poss
course. Nor will it be possible to do it even in an
stricted as Toyota City, but the geographic pattern of lean
tion in North America is already clear. The transplant a
OPerations (with the exception of NUMMI) are located within
300-mile radius in the American-Canadian Midwest, ~h
assembled in these plants a t first contained only &out 20
U.S. and Canadian content. but this figure rose steadily
60 Percent in 1990. and we expect it will reach 75 percent by th
late 1990s.
The supplier plants, some old and some new, a
located nearby, so parts can be shipped from supplier
bler in less than a day's drive. (Comparisons with japan
of the geographic concentration of suppliers can be quite
ing. Road congestion in Japan is so severe that suppliers located
within j0 kilometers [30 miles] of an assembly plant may actual1Y
need more time to deliver their parls than supI?liers 200 kilomefrom the Japanese a s s e m b l ~piants localed in
ters 1125
rural areas in the u.s.-Canadian Midwest.1
now
Honda, T
~ ~ i s s ~a nMazda,
, ~ and
~ Mitsubishi
~
,
~
r
o
c
~
established ~~~~h American product- and
operations as well. ~
~true to~ its conviction
d
~ about, cioil'g it
in one place, has located its engineering cenler at its Mdrysville,
located
the
Ohio, complex, &ile the other companies have
Detroit ares, ~
h reasons:
~
i They
~ want to be
quarters of U . ~ suppliers
.
and be able to recruit engineers easily
are growing rapidly, although it will be past
the Big
of the century before they approach the
the
Three American companies jn Detroit. However, they are
.,.hese
~
~
FIGURE 8.2
F N ~
Locafion
Assembiy Planis:
Honda
Marysvilie. OH
In Europe, the Japanese advance has been
slower, for
Alliston. ON
Mar da
Dlamond Star
CAMl
SIA
ASSEMBLY TOTAL
Engine Plants:
Honda
Nissan
Toyota
TOTAL ENGINES
Notes:
Flat Rock, MI
Bloomington, IL
Ingersoll, ON
Lafayette, IN
Anna. OH
57,000
76,000
200,000
150.000
140,000
4,000
40.000
200,000
15,000
14,000
15.000
35,000
80,000
22,000
50,000
50.000
Washington, UK
Barcelona, Spain
Swindon. UK
Smyrna, TN
Georgetown, KY
Longbridge, UK
Burnaston. UK
Hannover, Germany
Lisbon. Portugal
(1) Comrnenced operations in 1989.
(2) General Motors/Toyota ioint venture. Truck assembiy iine being added,
(3) Second assembly iine being added.
(4) Chrysler/hlitsubishi joint venture.
(5) General MotorslSuzuki joint venture.
( 6 ) Subarullsuzu joint venture.
Linares, Spain
Esztergom, Hungary
208,000
2O0.oo0
260,000
do0.000 (1)
200.000
(2)
(3)
14)
940,000
100.000 (5)
l00.000 (51
1,260.000
200,000
70,000
200,000
470,000
200.000
330,000
200.000
730.000
1989 Production from Ward's Automoiive Reporis. Capacity plans from company announc
rnenls.
Washington, UK
Swindon, UK
Shotion, UK
doing significant design and engineering. ~h~ body alterations
needed to create the Honda Accord coupe and station wagon from
lhe initial Honda Accord sedan were engineered at ~ ~ ~ ~ ~
and al1 the production dies were cut there as well. ~h~ coupe and
station wagon will be assembled exclusively at ~
~
~
for the~
~
whole world, with exports to Japan and Europe. ~
i ,jnn ~
Arbor, Michigan, engineering center is doing similar engineering
Notes:
~
~
~
i
i
(1)
(2)
l(3)
(4)
(5)
Production
by Rever lar
Honda. potentiai figure supposes a takeoverat Rever by Honda.
Toyota vehicles ascembled by Volkswagen.
joint
l venture
~ ,with General Motors.
Excluding the assembly of Land Rovers.
New plants under discussion.
l
~
l
~
~
'
~
osite. However, in either case, the fact remains
mn producers must either locate within the
the Japanese are doing in North America
t portion of the world motor-vehiele markei
volume producers seern to be doing in Norih
the E~~~~~~~~
ADVANTAGES OF GLOBAL ENTERPRISE
Besides the vital advantage of doing it all in one plac
point of sale, creating a top-to-bottom manufacturing syste
each of the world's major markets benefits a company in
waYs, compared with rivals trying to manufacture and ex
from a single region.
First, and most obvious, it provides protection from tra
barriers and currency shifts. For the company producing
site in one region, such as Jaguar in Britain
saab in swe
currencY
shifts can produce an export ~ i n d f ~ 1 l - example,
f~~
high profits these companies obtained in the united st
mid-1980sr when the dollar was strong in relarion to E~~~~~
currencies.
But disaster is equally likely. Between 1987 and 19
and Saab didn't get worse at making cars. In faet, our IM
assembl~plant survey showed a modest improvement
facturi*g productivity and product quality. M
~ both ~
Panies i~ltroducednew models that strengthened their pr
range. Yet during this period, a weaker currency in the
States,
company's primary export market, converted ja
Saab from high flyers to near-bankrupts. 'rhey wer
absorbed, respectively, by Ford and General ~ o t ~ ~ ~
with multiregional production bases,
For large companies that want to capture a substantial fra
tion of each regional market, the lesson of the 1 9 8 0 ~is
There is simply no substitute for within.the.region productio
The purchase of cars and tmcks accounts for about 15 percen
~and J ~
~
personal consumption in North America, E
su'b a 1arge number-$240 billion per year in the
America-that it's difficult to imagine offsetting
th
could
trade among regions when one ( J ~ ~ ~ ~
massive numbers of motor vehicles and the other regions consum
them.
TheexPerience of the 1980s also suggests that iftrade
do not arise to rebalance motor-vehiele trade, currency shifts will
These methods do have different consequences, ~
~
~
posed quotas on the import of finished units tend to
importrich as they raise prices lo ration demand, while currency
~
A second advantage [or the company developillg a multi~ t t o mproductiori system is rich product diverapter 5,the motor-vehicle market in E~iroPe,
nd Japan is progressively fragnlentillg, with no
oduction system can gain most economies
r volume per individual product compared
n, as we also saw in Chapter 5. sowever,
resumes that the variety of Products can be
erice on one large production line usi*g several
a
es of engine and transrnission from a large engine d a n t
ge transmission plant. SO companies with higher ~roctuction
eir products combined still have a competitive
long
as corporate management can deal with the
vantage,
mplexity, being big still rneans being better, and being bis in
in each of the
~
~ that ~companies
~ manufacture
,
~
~important,
~ consumers
~
l in the lthree regions
~
continue
demand different types of products and-this Point is ke~-to
t images to the same product. Consider the examG~~~~~
luxury cars sold as taxis in Germany to create a
,
le
base for their manufacturers, but sold in North America
much lower voiumes and much higher rices as
xuni goods. ~ i ~ i l a r l Honda
y,
has recently pockete* healthy
profi<s by exporting its Accord coupe, built and sold in the United
Statcs at high volume, as a much more luxurious, limited-volume
~
~
, the Japanese market.
product
for
~~~d~ seems to have been the first to see the ad\'antage of
bis approach. 11 plans during the 1990s to develop a set os
roducts unique to each major region. These will be produced
)
region to serve volume segments in that region. The
within
then export these products to other regions t" fill
arket ,,iches where it hopes their limited volume and exclusivit~
ill Permit charging higher prices.
1s~this approach
is ~carried ~to its logical
conclusion,
~
~
~
~
. the muktii
regional producer \vil1 have an intracompany product Porifolio
and trade flow as shown in Figure 8.4, where the maJoritY
~
206
THE MACHINE THAT CHANGED THE WORLD
and is met by the production system within each region and
S-regional trade is reasonably balanced.
A third advantage ¡he multir-egional producer can achieve
er the single-region producer is the sophistication managers
in through exposure to many different environments. Sophistition is subiective, of course, but in our interactions across the
with executives in al1 the major assembler and supplier
S, we've been struck by how much extra perspective managers
m trying to manufacture products in dilferent environ-
Post-Nationai Company
For example, we are convinced that one reason Ford has
performed better than General Motors in receni years is simply
that Ford has more of its production activities outside the Uniied
States and moves more personnel back and forth between iis
different international operating units. It's now rare to meet a
senior executive at Ford who hasn't spent years managing operations outside the United States.
By contrast, while GM has many foreign subsidiaries, it is
still common to meet GM executives who have punched their
ticket with a two-year tour at Opel in Germany or at GM of
Europe in Switzerland, but who otherwise haven't worked outside
the American Midwest. The wider exposure a¡ Ford produces a
higher leve1 of sophistication in operations management. Since
managers have heen exposed to radically different ways ol solving
problems, they also have the flexibility to think more creatively
about strategic issues facing the company. (Gaining the full benefit of international operations obviously requires a sophisticated
personnel system to rotate managers in the most productive lvays,
an issue we'll return to shortly.)
A fourth advantage ior the multiregional producer is protection against the regional cyclicality of the motor-vehicle market.
Motor vehicles are the foremost among what economists cal1
durable goods. With some patching, owners can almost always
get their cars to run a bit farther. So motor-vehicle sales in every
country tend to be more volatile than the general economy.
However, the world major markets don't go up and down at
exactly the same time; lor example, the Japanese market was
booming at the close of the eighties, while the American market
had gone soft. Thus, a company with a presence in al1 major
markets has more protection against cyclicality.
Establishing a global production system is particularly important for those U.S. companies relying predominantly un the
es~eciallycyclicai North American market. The Japanese
nies still se11 the majority of their cars in the .Tapanese
which, for reasons we'll examine in the next chapter, is muc
cyclical. They'll therefore find it easier to plow through the
automotive recession in North America and will cut pric
necessary to sustain smooth production at their new trans
facilities. By contrast, General Motors and Chqrsler largely
ate and se11 in the United States and Canada. Any fafl.off in
will force them to take money away from product-develop
activities and their foreign alliances to cover short-term opera
costs.
The product-development consequences will become ap
ent only in the mid-1990s, when these U.S. companies will pro
bly suffer further share losses. However, the slow market con
tions of 1989 and 1990 already have had some effects, chrys
reduced its share in Mitsubishi ~ o t o r from
s
24 to 12 percent a
GM reduced its equity stake in Isuzu from 44 to 38 percent, ~h
actions to raise cash are moving these companies in precisely t
wrong direction in terms of establishing a global producti
presente.
This fact becomes apparent when we consider the final
tage of developing a full-fledged production system in each of t
majar markets: Doing so denies competitors defended mark
from which to skim profits to use in competitive battles elsewhe
in the world.
The Japanese domestic market during the 1980~provides t
most striking example of what can happen when foreign co
nies cede a major regional market to domestic companies,
Western companies could have pushed hard to buy the weake
J a ~ a n e s ecompanies-Isuzu and Suzuki, in GM's case, and pe
haps
Mazda as well (Ford). This would no douht have led t
,,.
lnvestment friction" in Japan, where, as we have just seen, th
grouP equity structure effectively excludes foreigners unleSS the
grouP consciously decides to include them. However, this is a
issue that will have to be faced shortly in any case, and it woul
have been in the interest of the Western firms to oush - - a bar
----on this point.
Instead, they pushed for trade liberalization, so thev could
more easily export finished cars and parts to Japan. Liheralizin
Irade was an uphill stniggle-even in the absence of trade barr ers-because the Americans really had nothing to se11 that was
com~etitive,either on price or quality, in the Japanese auto
arket except a few novelty products, such as Cadillac limoues, the car of preferente for Japanese gangsters unti1 they
nsferred their loyalties to Mercedes in the 1980s.
~ ~ ~ the
~ Japanese
~ h i companies
l ~ , received a windfall from
rth American and European quotas. When the Japanese were
d they could sell only a fraction of the cars they had sold
reviously, they simply raised their prices untii cales fe11 to
ired leve]. ~~d they reaped huge profits in the process. In, the Western quotas are arguahly the biggest public-policy
st the Japanese auto industry has ever received-more useful
the Ministry of ~nternationalTrade and Industry (MITI! ever
in japan.Tlle Japanese companies used their profits to wage
arket+hare war in Japan, probably selling below cost in manY
and ensuring that Western importers would have little
there, even if there were no trade barriers at all.
cess
~
hin the
~ late
~ 1980~,
,
the situation reversed. The Japanese
panies used large profits from the booming domestic market
here the Japanese government did help by sharply reducing the
mmodity tax on car purchases) to underwrite their massive
estments in production facilities in North America and E u r o ~ e .
without fear of U.S. producers filing suit
could move
they were dumping or practising other trade retaliation,
ause their profits were not used to se11 cars abroad below
panese prices. Instead, their profits were used for capital investnts and new products, such as the Toyota Lexus LS400 and the
ssan Infiniti Q45, designed primarily for the U.S. and E u r o ~ e a n
~h~ U.S. and European companies, with no production Presce in Japan, made some profits themselves in the strong J a ~ a se market through a trickle of imports but missed most of the
ortunity, The failure to establish a manufacturing Presente in
or elsewhere in East Asia-to seriously challenge Toyota,
san, and Honda in their home market and take away this rich
lode-is surely one of the West's worst competitive lapses.
#ANRClNC THE GLOBAL ENTERPRISE
.ven the overwhelming evidence that a multiregional production
esence is now essential for success in the motor-vehiele indus, one question remains: how to manage a lean, global enter-
prise consisting of three top-to-bottom production comp]exe
the 1990s, and perhaps several more in tlie twenty-first cent
(for instante, in India for thc southern Asia market, in Brazil a
Argentina for the Latin American market, in Indonesia or Aust
lia Sor the Oceanic market, and evcn in South Africa-if
presenl niovement tolvard rejoining the ~vorldcommunity cont
ues-For the southern Africa region).
This is not a trivial management issue. Indeed, dynamic
effective management of global production organizations
largely defied the ingenuity of the automotive mass-produc
over nearly a century of trying.
The first auto company to pursuc a global man
strategy was Ford.' The present Ford Motor Company
in 1903 to manufacture the original Model A. By 1905,e
annuak production still totaled less than a thousand
Ford had established a Canadian manufacturing plant to
ble Ford cars Sor sale in Canada. In 1911, three yea
introduction of the Model T, Ford opened an assembly plant
Manchester in England. By 1926, Ford was operating as
plants in nineteen countries.
Hoi.vever, these steps hardly constituted serious internati0
alization. Ford's primary motivations werc to reduce shippi
costs-parts wcre cheaper to ship than finished units-and
surmount tariffs. Then as now, these were usually higher
finished units ihan on parts. Henry Ford made clear that al1 des'
and as much component fabrication as possible were to be
tained in Detroit. In addition, the foreign branch plants, as t
were called, were almost always under the management o
icails sent from Detroit.
This pattern continued throughout the 1920s.
one country after another erected trade barriers aiter
of the world economy in 1929, Ford was forced to go fart
built a fully integrated manufacturing complex at Dage
England in 1931 and a similar, though smaller complex, at c
logne in the same year. By the mid-1930s, these plants wer
producing practically al1 the parts for Ford's products.
radical from Henry Ford's perspective, they manufactu
product, the Model Y, not produced in the United States. This was
Henry Ford's belated ackno~vledgmentthat Europeans did not
want to drive American-style large cars.
We must remember, however, that the Model Y was designed
in Detroit, and many of the tools for its manufacture were made
ere as well. While English engineers suggested ways to accomcar to European tastes, the Model Y and al1 Ford
the 1930s were practically 100-percent America11
o n f y after the war did Ford of England and Ford of Germany
egin to hire their own product-development engineers, and not
ti1 1961, with the introduction oí the Ford Anglia, was a Ford
duct for the first time completely designed in a foreign coun4 This development occurred exactly fifty years after Ford
ropean assembly operations at Trafford Park, ManBy this point, the Ford Motor Company had turned 180
ees from its original practice. Where Henry Ford had deded 100-percent control of the product and ensured that al1
g decisions came from Detroit, Henry Ford 11 premarkable process of decentralization in which the
Ford of Europe shared no common ~ r o d u c t swith
etroit. It also had limited personnel transfers-that is, only a
ans in senior positions. It was in many ways a totaily
parate company in al1 respects save financial.
Because it recognized the emergence of a unified Western
rope before West German, French, or British companies didcoming the first "European" company in Europe-Ford of EUpe (established in 1967) became remarkably successful and was
key contributor to the survival of Ford in North America.
ssive loans from For-d of Europe tided Ford over during the
at North American auto depression of 1980 to 1982.
However, from the perspective oE senior management in Deoit, the evolution of a largely decentralized company was far
om ideal. By the 1970s, the company in North America had
loped a wide range of products smaller than the standardAmerican car of the 1950s. Many of these cars were identical
vera11 dimensions to the products develo~edseparately by
rd of Europe. It seemed only logical that global standardization
products in each size class would produce enormous savings in
evelopment costs and manufacturing economies.
Ford's first attempt to standardize on a global basis was the
Escort, introduced in 1979. A world design team was designated
to develop this car with contributions from al1 of Ford's global
operating companies. However, in thr process a curious thing
happened: The Europeans from Ford of Europe and the Americans
from North American Automotive Operations managed to specify
change after change in this "world" car to accommodate, respe
tively, European and American tastes and manufactLlringpref
erices. On launch day, the European and American Escorts,
though practically iildistinguishable in externa] appearan
only two parts-the ashtray and an instrument pa
brace.
In 1979, Ford bought a 25-percent stalte in Mazda in Jap
Because Mazda also malces a full range prciducts froni smal
lar% i t seeined logical to integrate some of ~~~d~~~prod
into the worldwide Ford p r ~ d u c t - ~ l a n n i nand
g
process.
For a start, Ford establishcd its own distribution channe
Japan (Autorama) and began seIIing restyled Mazda 121,323,
626 models there with "Ford" badges. These [riode]s are a]so
as Fords in many markets in Southeast Asia. A bit later F~
decided to import a restyled version of Mazda's small 121 de .
to the United states Srom Korea, where it is asseinbled by KI
sfl1all firm in which Ford and Mazda both hold a sma]l eq
stake. This model is sold under the name ~~~d ti^^.
BY the time the Ford-Mazda link was fully e s ~ a b l i s h ~itd ,wa
late to consider a joint design exercise for ~
~
(launched in 19851, but joint design was carried through on
new Mazda 323 and Ford Escort (launched in 1989 in
an
in 1990 in the United States). A similar cross-regional exerci
involving Ford of Europe and Ford North America (called CD
27) is now under way on the new Ford TempoiTopaz for the
American market and the Ford Sierra replacement in E
~d
in 1991.
Ford calls the process of joint design, with the lead role
assigned to either Mazda at Hiroshima ([or 323/Escort) or ~~~d
North America in Dearborn (for the next gel~e~ation
of large car
replace the Taurus/Sable) or Ford of Europe in the unit
Kingdom and Gerrriany (Eor TempoiSierra), "Centers of Respon
bility." Senior executives in the company have advocated thjs
a ~ ~ r o a cashthe only ivay to control spiraling development costs
for new ~ r o d u c t sat a time when a greater variety OS cars and
trucks is needed in every regional mar&.
However, thus far Su11 implementation of Centers o[ &spon.
sibility has eluded Ford's best eflorts. Ford of Europe argued that
the new 323~Escortwas too small Sor Europe and has pushed
ahead with its own design for launch at the same time,
in 1989 it introduced a new Fiesta model in the next smaller size
aster rejecting use ol. the Mazda 121 design (which was
ged toa small). Finally, executives in Europe are resistin~the
"f their large car (the Scorpio) in the Taurusisable
acement program on the ground that no single design can
sfy both American and European concumers in this class of
more, Mazda, while happy to act as lead designer on
~~~~~tproject, has continued to design its oTvn models for
size alid market classes-121, Miata, 626, and 929-and
e models continue to compete directly with Ford products in
ajor regional marlrets.
is sobering 10 realize that even with its limited pro,.;ress in
lizing design and production, Ford is still the clear leader
ng all companies, including the Japanese, in establishing
l as a truly global orgariization with design a1ld production
in the three major markets. By contrast, Chrysler has
ly a tiny manufacturing presence outside North America. conan agreement with Steyr in Austria 10 assemble 30,000
ing
rysler vans each year (beginning in 1991). General Motors has
trong presente in Europe and Brazil but continues to run these
rations as decentralized, stand-alone companies that hardly
lk
to ~
ils iqorth~ American
Finally,
the~ Euro~eanu
~ operations.
i
~
ned companies have either never started the globalization
ocess or, as we'l] see in a moment, made only halting ProgreSS
a few locations in developing countries.
~h~ J
~ by contrast,
~
are now
~
showing
~
an intention
~
to~
balize after strong initial reluctante and have had some initial
~ hurdles
~ ~ in the
, coming decade, as
~
~cess. H~ ~ ~\they~ face
, ~ huge
e will see in a moment.
EUROPEAN FAllURE TO GAlN A GLOBAL í'RESENCE
e E~~~~~~~industv now trails the Americans and the M===
globalization, as shown in Figure 8.5. When we consider the
riente of the Europeans, a fundamental axiom emerges: 1t is
establish lean production on a global basis when
sible
ave not mastered it at home. The case of Volkswagen Pros a good illustration.
I~ 1974, Volkswagen established an American assembly ~ l a n t
~
~
~
~pennsyjvania.
~
~ Its
~ objective
.
~ was
] to~establish
~
da ,
er.cost Anerican manufacturing base as the German mark
~
we can
Home Counlry
appreciate the full cost of Renault's setback
i1-i
thc
Local Region
Ford
General M o t o r s
Volkswagen Groop
Fiat (exci Iveco)
Renault (excl RVI)
PSA
Honda
Nissan
Mazda (incl Kia)
Toyota
Mitsubishi
Notes:
Exciudes double caunting of kits assembled abroad.
Loca' Regions: Americans-USA, Canada, and Mexico; EuropeansEEC, ERA, pol
Turkey. and Yugos!avia; J a p a n e s d a p a n , south Korea, Taiwan, ~ h ~ i l ~~ ~ ~d , i1
nesia. Phiiippines.
olfensive in North America. However, volltswagen knew noth
aboul lean production and staffed its U.S. plant with old-l
manufacturing managers lured away from General ~
~
ma1erializ.e. Equally damaging, the product adjustments
for the American market caused quality to tumble while al
revitalizing some of the worst mass-production plants in
America.
~
~
~
i
~
,
hen the Europeans master lean-produclion methods will tiley
in a position to revitalize their manufacturing presence in
t
~
~
~
,
IHE JAPRNESE ANO GLOBAL PRESENCE
o major business activity other than cars and motorcycles.
pletely atypical for Japanese companies, it also had little
est in the truck market, limiting its offerings lo a single
Given iis overlvhelming reliance on exports, whicli accou
for about 70 percent of its Japanese production, Honda deci
by the mid-1970s that it would be necessary to produce abro
Its vulnerability to currency shifts and trade barriers w
toa great if it didn't spread its manufacturing base. 11s U.S. a
assembly plan1 opened in 1982 but was initially only an assem
operation producing cars with perhaps 25 percent American m
ufacturing content versus 75 percent Japanese,
At the same time, Honda was searching for a manufactu
base in Europe. This was much liarder to develop, because H~
started selling in Europe considerably after Toyota an
and even af'ter Mitsubishi and Mazda. So it was last in
for the quotas established on Japanese imports
France and Italy by the early 1980s and had a very
bution network ir1 the more open markets, such as G
European sales of only 140,000 units in 1989, spread
models, Honda was in a weak position to move immediately t
full-size assembly operation.
lnstead, Honda pursued a stormy alliance with ~ o G~~v
initially a state-owned company but now part of private.s
Brilish Aerospace. After several licensing agreements in
Rever built Honda designs in England, this led to collabora
on the design of the model that became the Honda/Acura L~
and the Rever Sterling. Honda planned to se11 Legen
at Rever's U.K. Cowley plant to bolster its sales volume in E
However, it reportedly found the cars of unacceptable quali
even after rework at a new Honda-owned plant a t swindon
western England. So it quietly discontinued this effort short
after it began.
The next step was to design and produce jointly a new mid
car, the Honda ConcertoiRover 200. In 1989 Honda took a
Percent equity stake in Rover and provided a large amoun
manufdcturing assistance on the new ConcertoIRover asse
at Rever's Longbridge pfant near Birmingham. The product
launched in Europe in late 1989 and will be folfowed in 1992
newjoint ~ r o d u c tthe
, Syncro, produced at Honda's own assem
plant at Swindon, which Honda will open a t that time.
Honda has therefore moved painfully toward a EU
based manufacturing system, whose final form is yet to eme
through a complex collaboration with Rover. Meanwhile, in
United States and Canada, it has steadily expanded its asse,,,
plants at Marysville and East Liberty, Ohio, and Alliston, anta
city should reach 600,000 units by the end of 1990. Combined
its imports, Honda will probably pass Chrysler at this point
number three in North American passenger car sales.'
f greater interest Eor our purpose, Honda has steadily inased the ~ o r t hArnerican content of its cars by adding a
OOO.unit engine plant at Anna, Ohio, and a host of wllolly
rations nearby. lt also obtains a wide
m traditional Honda suppliers sr«m Jamerican transplants nearby and from
ed suppliers. While local-conteni calit will
tions are notoriously unreliable, Honda's claim
75-percent North American manufacturing value in its
ted states and Canadian assembled cars by 1992 is probably
far off the mark. ("Local content" is simply ihe postion of the
in the United States. For example, the engine n a d e in
the engine computes made in Japa* is
~
~
H~~ companies can add engineering value is a much more
eresting question when it comes to achieving global presente.
the lead among the Japanese transplants: in
tablishing an American engineering operation, both sur prodts and manufacturing processes. What's more, Honda is already
, the Accord coupe, which is styled and tooled
merica; has a second, the Accord station wagon, in Preparadesigning and engineering products from the
und up in North America by the end of the 1990s.
~ ~ w e v ewe
r , shouldn't undesestimate the scale of this
00 engineers in Ohio and Michigan by 1991
mingly a large number until one remembers that Ford and CM
sands of engineers in Detroit. Even taking into
ount our findings from Chapter 5 that Honda and 0 t h Japa~
e firms are likely to use engineers up to twice as efficiently as
Americans, Honda has a ion&way to go to implement LoP-tottom lean production in North America. This process took Ford
e. Honda is renowned for its abili~yto do
t we shouldn't underestimate the probiems
complete product-developmeni s ~ s t e m
a new continent.
if Honda can move very rapidly, we have to ask how the
pany's growing global operation will be managed. Honda's
t it will build an alliance of self-reliant
Japan, North America, and Western Eu1
THE MACHlNE THAT CHANGED THE WO
r"Pe2 and even in Latin America (Brazil) and southeast
(Thailand). The major regional companies are to conduct top
botiom design, engineering, and manufacture of products, ~h
are lo be sold primarily in their region "f manufacture,
limited volumes are also to be exportect to other
se
niche markets in a pattern similar to that of the hypothet
"post-national" company in Figure 8.4. 'rhe ~~~~~dcoupe,
bein& exported from the United States 10 Japan and
t
e x ~ o r t e dto Europe as well, is the first example this
But how are the regions to coordinate their activities?
wilI a world Honda personnel system loolc like? wil] the
jobs at headquarters still be reserved for J~~~~~~~nati
joining the company at age twcnty-two? H~~ long will it ta
bring about the alliance of self-reliant regions? ~h~~~ are
questions Honda must answer if it is to succeed
beenmi,,
truly global enterprise.
1
global production, supply, product development, technology
quisition, finance, and distribution.
~h~ central problem is people-ho~ to reward and motivate
ousands of individuals from many countries and cultures so
at they work in harmony. Unfortunately, the threc models
veloped so far for this enterprise are inadequate. The first is
treme centralization of decision-making al headquarters, a
arters a~mostinvariably located in the home countCV and
by nationals of the home country.
we've seen, this was Ford's approach from 1908 into the
os and is the approach of many Japanese companies moving
hore now. centralization produces bad decision-making. Much
orse from a political standpoint, it generales intense reseniment
other regions, as it soon becomes apparent that the most
mportant decisions are always reserved for headquarters and for
hose employees with the right passport.
~h~ commonly pursued alternative has been extreme decenrali7a+inn
subsidiaries, eacb developing its own
--... into regional
oducts, manufacturing systems, and career ladders in isolation
om the other regions. This was tbe position of Ford of Europe in
e 1 9 7 0 and
~ still describes GM of Europe. This hermetic division
,'
r
r i l i ~ t s in a narrow focus, ignores advantages of
-------oss.regional integration, and creates gilded cages for highly
&id national executives unable io rise any farther in their
-
SPEClFYlNC THE MULTIRECIONAL ENTERpRlSE
multiregional production system, have made considerable pr
gress, although neither ~vouldclaim i t has yet found th
solution. BYcomparison, however, the Europeans and Chrysle
[he United States haven't left the starting gate, and the rest of
JaPanese, including Toyota, substantially trail Honda. clearly, t
world and the auto industs. have a Iong way to
before mul
regional production is fully implemented. we'l] consi
challenge from a political perspective in our final chapter, H~
we look at the management challenge these companies face
layin&out the features of a truly global enterprise that can achie
multiregionai lean production in the lyyos.
Our goal is to specify the ideal enterprise in much
buyers of such craft-built cars as the Aston Martin
to spec
the car of their dreams. Unfortunately, no
dream machi
currently exists, SO we will create it: Multiregional ~~t~~~
The management challenge, we believe, is simple in
devise a form of enterprise ihat functions smoothly o,,
multiregional basis and gains the advantage of =lose contact wi
local markets and the presence as an insider in each of the maj
regions. At the same time, it must benefit from access to syste
,
-~
strategic alliances with independent partner firms from
ant on the last approacb, is the third model. ExamMitsubishi with Chrysler and General Motors with
talked of a
and Suzuki. (Indeed, Lee Iacocca has
bishi/Chrysler/European producer aiiiance, calling it Global
these arrangements leave the central question
and overall management unanswered. Given this
t, it is hardly surprising that most strategic alliances in the
(as differentiated from narrowly focused joint
NUMMI with specific, short-term objectives)
ave proved undynamic and unstable. The continuous bickering
d Mazda, GM and Isuzu, and ChrJ'sIer and Mithi suggest not that these arrangements need better manage, but that they are unmanageable except in perfectly Stabie
form we term the post-national. The key features of what we
Multiregional Motors are as follows:
An integruted, global pevsonnel system ihat pvomotes persa
COLlMtW ir1 the company as if nutionality did not a i
fin7
Achieving this goal obviously will require great attention to lear
ing languages and socialization and a willingness on the part
Younger personnel to work Sor inuch of their career outside
heme country. However, we already see evidente that youn
managers find career paths of this type altraetive.
We have mct numerous Japanese managers at the U.S. t
plants who look Sorward to long U.S. tours and
future as
ments in Europe. TJnlike older managers who often lack ]an
skills. they see this path both as an interesting way to live
surest routc to success in their company.
Similarly, Ford of Europe has recently had considerab
cess in recruiting European managers who do not expect or
to work in their home country and who anticipate serv
considerable periods in the United States as wel]. And we are
runnin&across a number of Americans eager to work in Japan,
A set o¡ mecl~anismsfor continuous, horizo~ztalinfo
flolv among n~antlfacturin~,
supply systetn.5, pvoduct del
technoiogy acquisition, and distriburion. The best way p
mechanisms in place is to develop strong shusa-led teams
~ r o d u c tdevelopment, which bring these skills together wit
clear objective.
In most Western companies, much activity is
Product planners work on products that never get the gre
massive amounts of staff waste time fighting fires. ~h~ best J
nese ComPanies, by contrast, believe strongly that if you a
working directly on a product actually heading for the mar
You aren't adding value. So involving as many employees
possible in development work and production is vital. compani
should keep their eyes on the product the consumer will buy.
Teams ~ o u l dstay together Sor the life of the product,
team members ivould then be rotated to other productment teams. quite possibly in other regions and even in
s~ecialties(for example, product planning, supplier coordina
marketing). In this way the key mechanism of inlormation
~ o u l dbe employees themselves as they travel among technic
specialties and across the regions of the company, E~~~~~~ wou
ay fresh a n d a broad network OS horizontal information ehannels
ould develop across the company.
pams
in ,lapan do stay together now, but members aren't
projects in ncw regions as a way to create a global
of horizontal knowledge arid to give every einpi~yeea sopklisated understanding of the world. (The question, os course, is
t whether this is a good idea in principie, but whether enouch
loyees wou]d find i t attractive.) As they travel through tl->e
pany and across regions, these managers also would create a
y culture-a largely implicit way of thinkiilg and
oing things that every organization needs to function well.
A nlecha,li.sin rol- coovdinuting the developmenf O¡ MeEv protialld fuci[iiuti~?gtheiv sule cis niche p r ~ O í i ~ lin~ f . ~
cts i12
ther ui<gio~~s-withoutproducing lowest-common denomiriator.
roducts. ~h~ logical way to accomplish this goal is to authorize
region to develop a full set of products for its regi«nal
arket. Other regions may order these products Sor cross shipproducts wherever demand warrants.
ent as
since MRM will ship products in rouglily equal volumes
al markets, it can largely ignore currency shifts:
ped in one direction are offset by higher profits
e other direction.
uickly. Or they search for trade protection.
MRM'S managers, who will have a long-term commitment to
wor]c~.c]asslean-production system in each major region, can
laxed, provided that one additional element of
e post.national enterprise is in place: internationalized financtodWts motor-vehicle companies have the bulk of
~~~t
their equity and borrowings in their home regiori and pay dividen& and loans in their home currency. So shifts in currency
lues are still a problcm, even if they have succeeded iu establishg a multiregional production system.
consideran American firm with dollar-based loans and divinds. A strengthening of the dollar could prove very damaging if
arned the bulk of its profits offshore-even though
he company's rnarket p s i t i o n and profitability in local currency
erms in al1 three regional markets was unchanged.
lnternationalizing corporate equity so that funds are r a i d
222
in each region in rough correspondence with sales volume a
manufacturing investment would largely eliminate this conce
Dividends could then be paid in the currency of the region
insulate the organization from currency shifts between regions
With these new approaches to personnel management, info
mation flows, product development, cross-regional trade, a
internationalized finance in place, it may be possible to create
MRM appropriate to the regional world of the 1990s. We be1
it is particularly important that motor-vehicle companies suc
MRM come into being, not just for commercial reasons but
cause of the emerging global political challenge. We will retur
this point in tlie final chapter.
e have now gone through the elements of
n production in the factory, in product development, in the
pply system, in the sales and service network, and in the
pothesized multi-regional lean enterprise. Our conclusion is
ple: Lean production is a superior way for humans to make
ings. It provides better products in wider variety at lower cost.
ually important, it provides more challenging and fulfilling
rk for employees at every level, from the factory to headquars. It follows that the xvhole world should adopt lean production,
d a s quickly as possible.
As with so much else, however, this is easier LO say than to do.
henever a fully developed set o€ institutions is firmly in placeis the case with mass production-and a new set of ideas arises
challenge the existing order, the transition from one way of
ing things to another is likely to prove quite painful. This is
rticularly true if the new ideas come from abroad and threaten
e existence of major institutions in many countries, in this case
e massive home-owned, mass-production motor-vehicle comnies. With help from their governments, these institutions may
sist change for decades or even overwhelm the new way of
So we are not certain that lean production will prevail. We do
lieve that the 1990s will te11 the tale. We are convinced that the
ances oF lean production prevailing depend critically on a wide
ublic understanding of its benefits and on prudent actions by
Id-fashioned mass-producers, by the ascendent lean producers,
nd by governments everywhere.
225
226
In the remainjng chapters, rve'll shift from analysis-wh
lean productiOn is and where it carne bm-to
prescription,
a vision of how the world can make the transition to
new
betcer WaY of making things with a minimum amount
pain and tension.
a
CONFUSION ABOUT OlfFUSlON
9
Between 1914 and 1924, Henry Ford's and ~ l f r e d
Sloan's industrial innovations destroyed a vigorous American industry, the craft-based motorvehicle business. During this ~ e r i o dthe
, number
of U.S. automobile companies fell from more than 100 to about a
dozen; of which three-Ford, General Motors, and Chryslerccounted for 90 percent of al1 sales.'
Yet there was no panic, no protest, no cal1 for government
tervention. True, a series of social critics questioned the new
type of factory life that mass production was introducing, but no
one called for the protection of the embattled craft producers.
The reason for the lack of resistance is not far to seek. Even as
Ford and Sloan were demolishing one industry they were creating
a second-the mass-production motor industry, and they were
doing it in the same city where craft production had flourished
most vigorously. The growth of this second industry ivas so dramatic that practically al1 the skilled workers of the craft-based
industry could find jobs building tools and doing other skilled
tasks to support the mass-production system. Indeed, until 1927,
when Model T sales coilapsed, Henry Ford faced the continua1
problem of finding enough skilled workers in the Detroit area lo
man his toolmaking operations. Meanwhile, the rapid growth of
227
carand truck sales, combined with continually falling prices
creati%'
hundreds of thousands of new unskilled jobs on
assembly line.
In additiol1, Ford and Sloan were Americans-homet
b o ~ seven-and
,
Henry Ford adroitly portrayed himself as a
hero bringing a high standard of living to the common
There was no suggestion of foreign menace i n tile triumph of in
production in Detroit.
No one has repeated Ford's and ~ l o a ncase
' ~ in suppla
One 1i.W of production method with another, ~
~ das ~soon
~ d ,
mass production began to move abroad
the united states
began to encounter i-esistance. This pattern is repeating it
loday as lean prod~ictiondisplaces mass production, ~h~ ba
Problem was and is that existing companies and workers us
older production techniques find it hard to adopt new
pioneered in other countries. The alternative method of di
techniques-the arrival of foreign coinpanies-tends qui
to set loose a nationalistic reaction in the countries where rhe
s t ~ l ecompanies are based. The result has "ften been a delay o
decades in substituting new production
for old ones,
MiSS PRODUCTloN ENtoURlTERS CRAFT PRODUCTIDN IN BRITA~N
In October 1911, Henry Ford opened an auto assembly plant
Trafford Park near Manchester, England." with the excepti
small assembly plant in Windsor, Ontario, just across the
River from bis Highland Park plant, this was ~
~first vent
~
d
abroad. Ford built the Trafford Park plant to overcome the li
lations o[ the transport of the day, but soon he neede
overcome trade barriers as well,
In 1915, Britain had abandoned free trade and adopted
McKenna tariff, which imposed a 25-percent tax on complete
autos that carne ti.om abroad. (Most of these imports
fro
lhe United States.1 Parts, b . contrast,
~
bore only a
tari
so foreign manufacturers had a strong inccntive to establish fina
assembly plants in England.
Initiah" evevthing went well at Trafford park, ~~~d dispatched a large number of American managers from ~~~~~i~
replicate exactly the mass-production system he was perfecting
'
ighland park. When workers were hired, they werc explicitly
that they would be handymen-that is, nene of their
if they had them, would be needed, and t h e ~would be
.lable to move from job to job in the assembly hall.' (lndeed,
manager at ~ ~ ~Parkf estimated
f ~ ~thdtd it took o n l ~
fivc i"
minutes to train a worker to do practically any of the assembly
in the plant,) ~h~ first powered assembly line was
nine months after the frst powercd line
eptember 1914,
complement
a reality at ~ i g h l a n dPark. By 1915,the
Ford assembly technology and techniques ufas in place at,
The implications of Ford's mass-production SYsteIn \vere uot
illed workers Ford was hiring for the b o d Y b ~ ' i l ~ ~ ~ g
~h~ upholstery department, for example, used s ~ e c i a l
to eliminate the skilled job of hand stuffins. Scat-cover
andardized and simplified. ln the b o d ~
sh«P, clampiminated the skilled panel beaters (\ghose
nts we recently encountered a t Aston Martin). A ~ a i n t - ~ ~ ~ ~ Y ~
tuted for the skills of the brush painter. The result
1913 was a strjlte that closed the body shop, as skilled
Ford's methods and argued for a return lo
ilders
illed work paid by the traditional piece-rate system." (Ford
,ken ranks with British employers and paid by the hour
afford Park, as he did in Detroit.)
B~~~~~~ worl<ers doing simplified tasks on the production
e could easily be replaced, and Ford, in any case, coulci s"PPIJ'
.shed car bodies from Detroit, the strike soon collapsecl. 11 was
ship car bodies lrom Detroit-the British tariff
e expensive
~ damage in transit drove costs up-but Ford coul" do so
xhausted their savings and gave in. BY 1915, no
as challenging Ford's system in the plant, and a Forcl manager
reported that productivity at Trafford Park was
producthat a t Highland Park.' Seemingly,
umphed in a new setting. Logically, it should 'Oon
e the dominan1 form of production in
perhaps in Europe as well.
However, this was not to prove the case. The reason was a
series f, events that make us very cautious about the rapid and
easy triumph of lean production in the 1990s.
killed craftsman highly suspicious o1 mass production. These
ont.line managers lobbied to retain traditional skills and ~ i e c e ate payment s"stenls, which made no sense in continuous-flow
worker's effort is paced by that of ez7crY
duction, where
THE TRlBULATlONS OF MASS PRODUCTION IN BRlTAlN
Ford's problems began in 1915 with an improbable event,
mission of his Peace Ship.6 Ford was an ardent isolationist:
United States should stay out of the First World War, he protes
To that end, he organized a trip to Europe aboard a &artered
10 broker peace privately between Germany and Britain, xn
sin, however. the public perception of Ford's motives was emp
ically different from Ford's own statements: H~ was universal
thought to be pro-German. The result was popular resistance
Ford products-many newspapers, for example, refused to
Ford's advertising-and a loss of morale among ~
~B~. ~
employees.
Energetic action by Ford's British managers offset some
the bad feeling, but Ford's problems soon multip]ied, ~h~ ener
horsepower taxes enacted after the war were particular
hard on Ford products. The horsepower tax in particular, wh
was proposed to the government by Ford's competitors and wh
favored
"long stroke" designs over ~ o r d short-stroke
'~
gine,
proved
a
crippiing
blow.
Ford's
Model
T,
envisioned
as
,,
d
F ~ ~English
~ \ ' plants
~
went backuJard
The performance
enormous
gull
developed
beiween
practice in
point where
f When
f ~Ford
~ abandoned
d
Trafford Parl<
and i n ~ ~ ~Park.
d established a full-fledged,top-to-bottoin manulaciuring s ~ s at ~
~~ ~ ~in
~ 1931,
l ~
the~gap
~ became
d ,~ even -*=ter.
h
reover, it persists to this d ~ .
With the ~~~d ~~t~~ company-the inventor of the new
tem and ¡he industrv leada-making such a poor sholving, ii's
' rdly ~
surprising that Ford's English competitors embraced m a s
roduction with only partial success.
~
~NDIJsTRIAL PILGRIMS: THE TRlP TU NlGHLRND P M K
the spring "f 1914 H~~~ Ford was actually turning out two
roducts at ~ i ~ hpark:
l ~Model
~ d T's and refurbished captairis of
quence of Ford's bad fortune was that Iiis factory often ran at
endless
streani
of
industrial pilgl-ims began to arrive
dustry,
fraction of its capacity and the company back in &troit seeme
ound 191 1 i n a flow that continued for forty ycars. (Indeed, lhe
to l0se interest in its performance.
grimages ended only with the visit o1 Eiji Tovoda in 1950.) The
Not sur~risingly,factory performance seemed to deteriora
in ~
~ Michigan,
~ cvntains
~ an extraordinab
~
rd
steadily. Nene of the English managers shared a conception
allery of pilgrims photographed with the master.
management that was compatible with mass production. ~ h ~ .
Thev range from ~
~ citroen
d (Citroen),
~
é
Louis Kenault (Reof a manufacturing career beginning on the shop floor with ha
nault), and ~
iAgnelli ~
(Fiat), to ~nameíess~Russians~ arlxious~
on management was unattractive to middle-class ~
~
~
l
i to add
~ mass-production
h
~
~
techniques to Lenin's lormula of "So"iets
who emerged from an educationai system that steered th
electrification equals ~ommunism."(Leni11 later amended
toward the civil service, the law, and 0 t h types
"soviets plus prussian railway administratiun plus
high.le
this formula
administration. They didn't want to get invofved with the nit
~~~~i~~~ industrial organization equals ~ocialism.")~
A partic"gritty of running anything. Rather, they wanted lo delegate op
striking photo taken in 1921 captures Charlie Chaplin and
ational details, just as they did with the ~
~
~
i
~
~~~~din .smiling mutual admiration alongside the Highland
iry
In addition, British managers were persuaded that ~ ~ ~ l j ~ h - k assemb]y ]ine, a t a time when Ford was still perceived as a
men, with a long experience of craft working, would not tolerate
worker for the masses rather than the e n e m ~of labor.'
Ford's methods. For a short period perhaps-under the whip hand
wil!iam ~ ~ ~founder
~ i ofsthe, Oxford Motor ConlpanY
of the American managers-but certainly not over the Iong term,
subsidiary MG), and Herbert Austin, founder of the Austin
Consequently, managing the shop floor soon became, by deMotor company,
M,ere among these pilgrims. After Morris visited
fault, the responsibility of the shop steward, who typicaIly was a
~ i ~ h park
l ~ in~ 1914,
d he returned to England determined
van'[
~
i
willing
able to operate a Ford-style mass-production syste
Ouite incredibl~,Morris continued to pay
bis workers
ay
motivating his workers.
the finished product.
blocks and transmission cases along rails from machine to
chine. Then each machine would work more or less
a given task. (Previously, the machines had
pick the bodies up . . . and jump the pegs [to move the
faster than the assembly line] and we'd make it up to double time
and a half, which was about five pounds a week, and this is a lo'
of money in them days."'
The idea f, workers running down the line carrying autornohiles faster than the conveyor seems comical toda^. And the
system must have had horrendous consequences for the quality
but Austin c o u ~ dsee no other way manage. As
the finished
managers argued, in defense of the piecework
one
bis senior
system: -some
form of extra wage íthe bonusl must be paid
pay
to a man if he is expected lo work harder. The only alternative
task
to the Ford system and ;nsist
to pay a high wage
may
achievement. . . . The daily task system at fixed wages
perh,ps, be workable in American . . . factories, but the necessav
. , . dri,,ing works policy ~ o u l not
d be acceptable either lo
Labour or ~
~
~ (The English
~
~ term ~for factory
~
~
,,works,~so a "driving works policy" is simply machine pacing
with standard day rates.)"
The consequence of this hybrid system, now called lhe British
system of mass production,~was that British plants, including
~
234
235
an economic miracle of the 1950s and 1960s was nothin¿? more
of General Motors and Ford, never matched the produclivit
an
a belated embrace of mass production. Volkswagen built
q u a l i t ~of U.S. plants. Sndeed, it was not unti] the financi
lfsburg
as the world's largest car plant under a single l-oof,and
crisis of 1980, sixty-seven years aster the introduction of
enault
and
~i~~ responded with Flins and Mirafiori, al1 plants
powered assembly line at Highland Park, that R~~~~ (former
British Leyland), the successor to the merged ti^ and
ncluded in our survey reported in ChaPter 4.
the mid.1960s, continental Europe had finally mastered
comPanies, finally adopted standard hourly rates and explicit
techniques
~~~t~as Eiji Toyoda
and Taiichi
Ohno ~were
set out to match tlie productivity o[ the ~
~ ( ~ ~ i~ ~ i ~ ~
i
~
~
Le~land
ving
beyond
them)
and
began
to
challenge
Detroit
in exPort
nationalized in 1975. B." 1979, i t was hemorrhagin
the
same
time,
the
Americans
were
investin%
red ink, ami a new management was insta]led with
rkets,,Z
instm
ely in E~~~~~and had developed complete Production
tions lo make the company efficient or clase i t down,) B,, th
pment and supplier systems on a Europe-wide !mis. Tlle Process
lime, o í course, American-style mass production was alread
production with mass production had been
under siege rrom Japanese-inspired lean production.
replacing
completed, but it had taken fifty Years.
lhose
,
I'fWJUtTlON IN CONTINENTAL EUROPE
~RODUCTION
ENGOUNTERS MASS Pi@DUGTlON
The French, German, and Italian experience with mass produclion was a variation on the English ihemc, with the difference
lhat the Americans had a harder time spreading their hemegrown system through direct investment. citroen,~
~and
~
Agnelli,10 cite the three industrialists most taken with the mass.
production ConcePt, struggled through the 1920~and 1930s to
implement the idea in chaotic economic and political conditions,
Their problem stemmed partly Srom resistance by craftspeople,
but also from the lack of a stable domestic market as E~~~~~~~
economies careened from hyperinflation lo depression,
Ford tried to lead through example with investments col"gne in
Germany and Poissy near par;s, and GM bought the
tiny Gerrnan producer ~ p e in
l 1 925. Horvever, ~ t a l yfirmly c]oscd
jts door lo both companies. What was more, the need
~~~dand
GM t" Produce practically every part ior e v e i car within each
EuroPean country, due to trade barriers within E~~~~~and across
the Atlantic, pushed up costs, restricted the size
the market,
and, in general, retarded the spread "f mass production, As
EuroPe plunged into war once more at the end of the ]93Os, the
progreSS of mass production had been quite limited. sn
the
failure of the European economy to grow was one
thc
ing causes of the war. That is, because mass production hadnrt
progressed, the European economy stagnated, creating thc conditions that helped lead to war,
A[ker the war, change became very rapid, ~~~h "f the E
~ ~
~
We have paid careful attention to the substitution of mass Production sor craft production because of the perspective jt gives On the
~
~
l of superimposing
~
,
current
challenge
lean production
production. lnfact, the new challenge seems much greater.
In Europe in the 1920s, the craft-based auto industry was
quite small, ~h~ substitution of mass production, had it
ceeded, would cerlain]y have increased employment spectacularly, as it did when mass production finally arrive* in mature
form in the 1950~.Nevertheless, the threat 0f foreign domination
(by the ~
~was so frightening
~
and~ the mismatch
i
between
~
existing institutions and conceptions (such as English notions
management and continental notions of skilled work) so great
that Eufope sealed itself off rather than adaPt.
In the 1990s, the fear of foreign domination (this time by lhe
Japanese) will surely prove as great. tlowever, the mature nature
of the motor-vehicle market in North America and EuroPe, 'Oupled with the efficiency gains inherent in lean production,
that no painless solution can emerge. AS lean production
mass production but tiie same number of cars and trucks are built
each year, many jobs will disappear.
Whatrs more, the ,-unent Western automotive work force is in
precisely the opposite position of craft workers in I 9 l 3 . The
introduction
mass production created new jobs for
worklhe new
ers-these
workers
made
the
production
tools
needed
-
~
236
'ystem.
contrast, lean production displaces asmies of ma
production workers who by tlie nature of this system have
skills and no place to go.
THREAT
THE MEW ~N~uSTR~RL
PILGRGRIMS: THE TRlP 18 HlROSHlMR RND TOYOIfi
IlOflZoN: INITIAl MISPERGEPTIIINS O[ LEAN PRU~~UC~ION
rkers union, rcversing thc steps Ei.ii To~odatook in 1950.
In 1980, the Ford Motor ~ o m p a n ysuffered w h a ~turne' out
Anything that is new is likely to be misunderstood, typically
attempts to explain the new phenornenon ir1 ter?ns of lraditio
be a very timelo crisis. The company began to lose
categories and causes. So as the industrial revolution ~~~~d~an
Ohno had ws-oughl
to make itself fe]^ abroad, throug
exPorts of finished units, what they had achieved was widel
misinterpreted.
One popular explanation in the 1970s of the japanese
success
was s i m ~ that
l ~ J a ~ a n e s ewagcs were Iower, an explanation that
Iits n i c e l ~into estahlished theories of international trade, .4 secOnd ex~lanation
was sumrned up by the
"papan,I
~This~
,
~ They decided to go to Japan to see for themsel\'es. a .iourney
t h e o r ~attributed Japanese success to its governmentss proteclion
that became feasible by ~ o r d ' purchasing
s
24 percent os Mazda in
1979, This meant that senios Ford executives and FordPs UAW
of the domestic market and its financia1 support for J~~~~~~~car
leadership could gain full access to Mazda's main production
companies, through tax breaks anci 10%. interest rates, as they
complex in ~
i
~ and ~determine
~
h for themselves
i
~
~why Ford
tried t" tar&et export markets. A third explanation was high tech,
notably the widespread adoption of robots in the factory, Towas rakjng a beating in international competition.
connection,
gelher, these made the emergente of Japan understandable bur
Ford had a second stroke of luck in its
also sinister-artificia~~ylow wages combined witll go,,ernment
~~~d~ itself had experienced a timely crisis in 1974.
notably
Tlle failure of its technology-driven product strateg~-based
s"PPort (for example the 1970s tax laws promoting the
its use O[ the fuel.hungry Wankel enginc-caused Mazda t" tralisinstallation of roboticst-to beat the westcrn mass.producers at
their own game.
form its production complcx at Hiroshima into a faithful copy
Toyota,s lean-production sy"t"m a t Toyota City. If the Ford
What's more. there were elements of' truth in each of lhese
execulives had visited ~iiroshimain 1973 rather lhan 19812
explanations. Japanese wages were substantia]ly lower than
American wages prior to the currency shifts of
well have reached the wrong conclusions.
1970~.The
they
Japanese government's efforts to protect the domestic market and
After several weeks in Hiroshima, followed by months
the ~~~d executives and Ford's UAW leadcrs
domestic ownership were absolutely esr;ential to the initial
careful
discovered
the
answer
to Japanese success: lean produciion. spcgrowth of the JaPanese industry. And the leve1 automation
on
cifica~ly,they four,d lhat Mazda could build its 323 model with
average in JaPan by the early 1980s was higher than i n the west,
what these explanations could no1 explain was horv tile J~~~~~~~
only 60 percenl of the effortFord necded to manufacture its
in the same market segment. Moreover, Mazda made man"
'OrnPanies continued to advance in the 1 9 8 0 ~despite currency
manii[acturing errors in doing so. Equally striking,
shifts and a massive moirement of operations offshore, where ~ 1 ~ 1
p r o d ~ much
~ t ~ more rapidly and with leSS
was os little .&
]h
e'
Nor did they explain why the japanese
firms
gai*ed ma.ior benefits from automation while ~
~firms often
~
t
~could develo,,
~
~
~
worked
much
more smoothly wilh its sup.
effort than Ford and
lo Vend more than they saved. Deeper explanations of
these rnYstefies required an undesstanding o[ lean production,
1982 was threatening the survival of the company-was breakin
the l"gjam of ofd thinking and entrenched interests, suddenl
e m ~ l o ~ eaetsal1 levels of the company were
to stop thinki
about how to advance their careers or the interests of th
department and to start thinking about how to save the compa
This situation is the very definition of a creative crisis, and
news from the pilgrims to Hiroshima carne at lust the rjght
During the 1980s Ford was able to implement many elements o
lean production and the resuits soon showed up in the market
place.
Chryslerr meanwhile, was in much deeper trouble than F~
Or CM and was already a ward of the U.S. government, w h
failed to learn much about its real problems in a time of cr
despite its equity tie-in and access to ~ i t ~ ~ b iis~ ah trag
i ,
mystery.
General Motors' experience was quite different from ~~~d~
The comPanY was also represented along the pilgrimage route,
but ~ n t i recent
f
years it facked the crisis needed in any mass.
production ComPanY to take the lessons of lean production to
heart. GM was rich in 1980. Although it lost $1 billion in 1982, it
had little debt and was, by far, the world's largest company,
It dealt with its problems mainly by retreatjng from market
segment after market segment and attempting dramatic feaps in
productivity through the introduction of al1 available new production technolog~when it introduced new models
as ~ ~ - 1 0 ,
No One com~lainedas the Japanese moved in to fill the competilive gap unti1 quite recently, when institutional in\,estors started
becoming nervous that GM was slowly liquidating itself.
In the 1 9 8 0 ~
GM's
~ primary means of education was through
the Planning process for its joint venture with ~~~~t~in california. As this plan became a real possibility in 1983, senior GM
executives spent a lot of time at Toyota City.
jack smith, now
GM's vice-chairman, noted later, " ~ was
t
the first time we
kmci a clear understanding of how they ran. . , . ~h~ data
productivityl was just unbelievable,"ia
As we showed in Chapter 4, the NUMMI joint venture was
extraordinar~success. However, transferring the lessons learned
throughout the vasi General Motors organization has proved hard
work. The fundamental problem is that making the transition
from mass ~roductionlo lean production changes the job of every
worker and everY manager. What's more, in the absence of market
growth, many jobs are eliminated. ~ i n c eGM didntt face a crisis
con
the 1 9 x 0 ~and failed to find any opportunities for groxvth, i t
was simply 110~able to face up to the challenge.
For the same reason, the European motor-vehiele companies
have been represented only modestly along the pilgrimage 'Oute
o T~~~~~city and lean production. The Euro!>ean auto nlarket
as vigorouc in the last half of the 1980s, setting a new
rd each year from 1985 through 1989, and Japanese coinpetiwas contained through formal irade barriers and a \yelter oE
entlemen8sagreements,~'AS a result, the European companies
ad little external pressure on them to cliange. As meilliolled, the
ost notable movement toward lean production in Eur"Pe 'ms
by European companies but by the American ComPan?.,Ford,
ich attempted apply in Europe what i t had learned i l l J a P n .
An experience of our group pert'ectly summarizes this situation, ln 1982, &ile visiting a French auto asseinbly plan? in lile
Paris ares, we encountel-ed a young engineer. Fie had just returned
a year-and-a-half exchange visil ir1 a Japanese
the plant
car company in japan. tie was bubbling over with enthusiasm
about the contrast between lean production, as he had discovered
it almost by accident in Japan, and the mass-production Practices
f,
bis own company. H~ was eager to introduce lean-!?rc>duction
techniques as quickly as possible. His main concerri was w h ~ r eLo
begin
holv
capture the attention of senior mand&ement.
Our discussion was cut short by a remarkable event-a violent industrial action involving two factions o[ the North African
guest worbers who held practically al1 the production jobs in Ihc
plant. ~h~~~ workers were represented by two sepal-ate ulliolls
aIid werc embroiled in a dispute over work rules. As tbc tcnsion
belween the two factions grew toward a confrontation in n'hich a
fin&& vehieles \vere vandalized, thc plant
large number
managers advised oul- team to leave. We wished the voung engiluc]< in implemcnting lean productioil as we
neer the best
hurriedly departed.
1989, quite by accident, we ~~~~~~~~~ed lhe
In the fall
one
of the provincial plants of bis comPanY
same engineer
where he was now head "f manuiacturing We asked what had
become of bis el.forts lo institute lean production. For a moment
he looked puzzled, but then he remembered our initial encuuntcr
and gave us a remarkable reinterpretation o1 events: The
problem, he had concluded, was the guest workers in the French
~n the provinces, ho~vever,guest
auto plants in the paris
issue. M1 the workers were French, a
workers were not
240
THE MACHINE THAT CHANGED THE WORLD
cooperation prevailed, and he would stack up his current plant
against any in the world.
We had considerable difficulty in continuing the conversation
at this point, because the survey we had just eompleted showed
that his plant takes three times the effort and makes three times
as many errors as tbe best lean-production plants in Japan in
making a comparable product. What's more, the amount of space
and the inventory lcvels in his plant were severa1 times the
Japanese level, and the French plant was focused on a single
product in one body style on each of its production lines.
In short, sincc his company hadn't faced a challenge from a
lean producer, he wasn't able to initiate the ehange in mind-set
needed to implement lean production. The young bearer of the
message had returned from the pilgrimage and fitted himself into
the familiar industrial landscape of mass production. We felt a
profound sense of gloom as we left the plant.
BENCUMARKINC THE PAIH 1 0 LEAN PRODUCTION
We in the IMVP have been pilgrims ourselves, first to the best
lean-production facilities-al1 in Japan until very recently-and
then back to the strongholds of mass production in North America
and Europe. We believe we have traveled farther and made more
comparisons than anyone else, either inside or outside the motorvehicle industry. So, where do we stand along the path to global
diffusion in lean production? And what must happen for the
whole world to embrace this system?
Remember that as a practica1 matter there are only two ways
for lean production to diffuse across the world. The Japanese lean
producers can spread it by building plants and taking over eompanies abroad, or the American and European mass-producers
can adopt it on their own. Which of these method uroves dominant will have profound implications for the world economy in
this decade.
DlFFUSlON THROUGH IAPANESE INVESTMEWT IN NORTH AMERICA
Japan's move offshore began as a trickle in the 1960s. Its first
major initiative was Nissan's engine and assembly plant in Mex-
ico in 1966. Not much else happencd for a long time-unless you
count extremely low-volume assembly plants ("kit plants," in car
talk), generally run by licensees rather than the Japanese company itself, in protected developing-co~intrymarkets. For example, in 1966 when the Brazilian government prohibited further
imports of complete vehicles, Toyota licensed a local Brazilian
company to assemble kits of parts for its Land Cruiser utility
vehicle.
Honda made Japan's iirst serious foreign investmcnt with its
Marysville, Ohio, complex, which began assembly in 1982. Oncc
one company was firmly committed offshore-and as it became
apparent that shifting currencies and persisten1 trade barriers
(for example, the Voluntary Restraint Agrcement on Japanese
finished cars entering the United States) made foreign investment
inevitable-al1 the Japanese companies rushed to follow Honda's
lead to North America.
The large number of Japanese auto companies (eleven) and
the intensity of their rivalry has led to an extraordinary investment boom, as shown in Figure 8.2 in the last chapter.
The assembly plants came first, followed by engine plants,
and now by a wide variety of parts plants. What's more, the
investment flow is still broadening. Honda, Nissan, and Toyota
have announced plans to design and engineer complete vehicles
in North America by the late 1990s. With that step, they'll complete the process of constructing a top-to-bottom manufaciuring
system. The other Japanese companies are sure to follow.
The speed and scale of this process are truly extraordinary.
Indeed, nothing like it has ever occurred in industrial history. In
effect, between 1982 and 1992 the Japanese will have built in the
U.S. Midwest an auto industry larger than that of Britain or ltaly
or Spain and almost the size of the French industry. By the late
1990s, the Japanese companies will account for at least a third of
North American automobile production capacity-perhaps much
more-and have the ability to design and manufacture entire
vehicles in a wholly foreign culture 7,000 miles from their origins.
What's more, politics permitting, these investments will continue until the American companies revitalize their operations
and stand their ground in the marketplace or are eliminated.
By contrast. Ford established an initial assembly plant in
Europe in 1911, added full manufacturing at two sites-Dagenham, England, and Cologne, Germany-in 1931, and completed
the process with a full product-development team in 1961. lt took
fifty Years for Ford to accomplish what the Japanese may
in fifteen years. General Motors was no quicker, lt bought the
company in Germany in 1925 but only moved to high-Sca
challenge" in which ihe American multinationals were seen
threatening to take over ihe entire Europea* motor induStry,i7
sized in an interview, "we believe that our production system,
with its many nuances, can be learned by anyone . . . but it takes
ten years of practice under expert guidance."
one accepts this manager's estimate of the time and Personit takes to transfer lean production-and we do--ir
follows
IAPANESE DOES NOT EUUAt LEAN
tn the excitement over the transplants, many people seem t
forget a point we stressed in Chapter 4: that al1 transplants i
North America do not perform at the same level. ~h~ best per.
[orming plant. Company Y, took 18.8 hours to perform our stan5
dard assemblv tasks on our standard car and needed
square leel of factory space per year per car. A competing transplant located nearby, Company Z, needed 23.4 hours per car and
used more than 13 square feet of factory space, by lar the least
efficicnt use of space in our entire world sample.
Both of these plants are Japanese, but one is much leaner
than the other. What explains this very substantial difference in
performance?
One reason: Company Z is not so proficient at lean production
in Japan. Its plants there also trail the performance of company
Y's. Again, %vemust stress that "lean" does not equal "
J
~
While average Japanese performance is very impressive, a few
JaPanese companies seem to have been inspired more by H~~~~
Ford than Taiichi Ol-ino, while a fcw companjes in ihe westironically. the Ford Motor Company is the best example-have
greatly modificd their factories and come elose to leanness in the
1980s.
A second reason for the difference in performance between
the best and worst transplants is that Company z delegated most
of
operational aspects of its plant, jncluding desigil and
layout, to Americans lured away from Detroit. This approach is
~
~significant.
~ foreign
~ governments
l
l may ~slow the efforts
of the japanese
companies by piacing restraints on the number of
~
expatriate managers they allow to work in their countries. The
U.S. government, for example, has taken an increasingly stringent
line on Japanese nationals managing the transplants, aPparently
from the conviction that the purpose of transplants is t0 create
~
~
~
~
~
,
~
obs for Americans. So it is naive to assume that lean production
can be transplanted instantly by the Japanese-just as it is naive
THE MACHINE
CONFUSlON ABOUT DlFFUSlON
DlFFUSlON THROUGN LEARNlNG BY TNE AMERIFAN FlRMS
But what o i the Americans? Where do they stand a]on
lransitional path lo lean production? Without question, the
indus1r-y as a ~vholeis gerting better a t factor-y oper.ations,
company has improved considerably. I-Io\velier, GM and chry
have improvcd their operations Iargeiy by simply closing
worst plants, like GkI Framingham, rather than by improvi
e v e v plant. The Chrysler St. Louis 1 assembly plant illustra
this process.
St. Louis 1 has been assembling 210,000 ~
o ~~~t~~~
d
~an ~
Chrysler LeBaron models with a work force of 3,400. ~h~ be
Japanese transplants can assemble the same number "f cars wi
2,100 workers. Chrysler and its union faced a
,-hoic
Coni'ert from mass 10 lean production ivhile displacing 1
workers, or close down altogether. Neither the company nor the
union found a way to make the transition to lean production,
the plant will close at the beginning ol the 199 1 modcl year.
This outcome has recurred repeatedly at GM and chry
over the past three years, as shown in Figure 9.1. The two compa
nies have together shut down nine North American plants, while
fully converting none to lean production.'" As this process contin.
ues. GM and Chrysler are enveloped in an ever-deepening sense
of gloom in which a slow retreat never quite seems to trigger
the crisis that might shake free the logjam o i outda& mana@.
ment thinking and union-company relationships and lead to
revitaiization.
Ford. as we've seen, has made good use of its 1981 crisis and
Lhe ~ilgrimagelo Hiroshima and found ways tu equal the productivity of the transplants. We are concerned, however, 1hat its
plant-level performance is best when it can focus its factories on
a single model witli only a few options. In plants with a complex
model mix, Ford's performance is much Iess impressive, so the
comPallY has traveled only part of the way down the path to
leanness even in the factory. Neveitheless, Ford llas made a bo]d
start and has bought time to perfect its own version of lean
production.
As we saw in Chapter 6, the Americans have begun to ration.
alize their supplier system. The number of suppliers to each
company has been reduced dramatically and the attitude toward
Year Clnsed
Detroit, Michigan
Norwood, Ohio
CapaciQ
Pontiac. Michi~an
1987
1987
1988
1988
1988
1989
1990
1990
1990
1990
212.000
250,000
250,000
300,000
100,000
200,000
200.000
230,000
210,000
54,000
Spring Hill, Tennessee
1990
250,000
Leeds, Missouri
Kenosha, Wisconsin
Pontiac. Michigan
Framingham. Massachusetts
Lakewood, Georgia
Detroit, Michi~an
St. Louis, Missouri
1,756,000
we also found e]ear signs of the intent to lnove to$vard leanness in the ares p o d u c t development. ~nfortunately,no prod.
uct launched lo date has benefitted from a truly lean development
process, an,jone must wait until a product is fully develo~edand
intro,juced in the market before drawing conclusions about its
development. program improvement~instituted in 1990 will be
clearly apparent only if the new model appears in 1993 1994that is, in three to three and a half years rather than in the typical
five-and with a greatly reduced leve1 of engineering office. Howwe have seen no clear evidence so far that the A*erican
companies can elose the gap with the best Japanese companies in
terms of development time and effort, only that they can
ically shrink traditional effort levels and time frarnes even as the
best Japanese companies continue to shrink theirs.
~i~~ years ago, the Japanese auto makers considered fortytwo months a satisfactory development pace. Today, the best
companies are ta]king about twenty-four months as a reasonable
target. soleanness continues to progress and the trailing wes1e1-n
mass.producers will need to move quickly indeed to catch
KRN PROOUCTION I N THE SEA OF CYCLlCRLlTY
From une perspective, the transition to lean producti
ceeding with remarkable speed and smoothness-the tra
have shown that lean production can thrive in North A
and some of the American companies show signs of maste
new system as well. What's more, despite widespread predic
of an overcapacity crisis, the advance of the transplants has t
far been almost perfectly synchronized with the retreat of
American-owned firms. Between 1987 and 1990 imports f
Japan of finished units, cars plus light trucks, declined by abo
million units, and 2 million units of American-owned
were retired. A t the same time, 2.5 miIIion units of
transplant capacity were added. Thus declining impo
retired capacity exceeded new transplant capacity by 500
units and actual capacity utilization in 1990 was only slig
belo~vthe leve1 of 1987, the drop-off being due to lower
1990 compared with 1987.19
On another level, however, many problems still must b
overcome if North America is to avoid the European fate of th
1920s, in which reforms in production were deferred fo
ation. Many of these difficulties are interna1 to the pr
system itself, while others are political; some are both. ~h~
include:
The cyclical pattern of the U.S. motor vehicle market, whic
is incompatible with lean production.
North American notions of careers, which are also incom
patible with lean production.
The fact that the rapid decline of U.S. and Canadian-owne
companies that many regard as national institutions i
likely to prove toa much for politicians and the genera
public to accept.
westerners are
resigned to tlie idea o[ the business cycle. Likc
gravity, itrs simply tjlere, although no one quite knows w h ~ No
.
one likes it, and cures have frequently becn proposed, the most
recent being Kcynesian macrocconomic management. To dale
none has worked.
* us Sales
--s;-
Japan Sales
m
-
.-o
..
5
$0
m
m
cc
"3
Let's take a closer look at each of these problems.
Yeai
Source: United States sales: Motor Vehicle Manufacturers Association. Motor Vehicie F c t s and
Figures. Japanese sales: Japan Automobile Manufacturers Association. Motor Vehicle Statlsfics o f Japan.
248
CONFUSION ABOUT DIFFUSION
FIGURE 9.3
--
Cyclicality of Motor Vehicle Production,
United States Gompared with Japan, 1946-1989
Figure 9.4 shows the consequences for employment in the
U.S. auto industry. (Note that employment of salaried profession21s has been much more stable than that of hourly production
workers constituting the bulk of the work force.)
The problem with the American pattern is that it is extremely
corrosive to the vital personal relationships at the core of any
~roductionprocess. Mass-production workers are
no illusions that their employer is going to stand at their side through
thick and thin. Indeed, the most important function of mass.
production unions is to bargain for seniority rights and for layoff
com~ensationfor those chucked over the side
the company
ship. Similarly, suppliers to the mass-production assembler comPanies are under no illusions about a shared destiny, when times
are bad, it's each company for itself. And the suppliers, in tuni,
jettison their own workers and subcontractors. AS no&, the
consequence is a distinct lack of commitment on the part of
workers and suppliers.
Lean production, by contrast. is inherently a system ofrecip-
roeal obligation. worlters share a fate with their em~loyer,s"ppliers share a fate with the assembler. When the SYstem works
t"
properly, it generales a willingness to participate actively
initiate the continuous improvements that are at ¡he verY hezirt of
leanness.
But can this system work in a cyclical economy? As figures
9.2 and 9.3 show, the issue never arose in Japan because neither
¡he domestic auto rnarket nor domestic production is cyclical. (As
the low-cost, high.quality global producer of automobiles untik
ves. recently, the Japanese domestic industry has always been
able to ,,low through slumps in export markets by cuttillg margins,) lndeed, the largest contraction in production in JaPan Over
the past forty years is smaller than the smallest contraction in
North America.
what happens as lean producers, whcther Japanese or American, encounter heavy seas in North America (and, Lo a lesser
General
~ ~~ilotors
) ?executive gave us one
degree, in E ~ ~ A ~
THE MACHlNE THAT CHANGED THE WORLD
answer, ~vhenhe looked a t a version of figures 9.2 and 9.3 duri
an interview: "When the Japancse [meaning lean] producers e
counter these gigantic market waves, they will quickly become a
mediocre as we are. They will llave to start hiring and firin
workcrs along with suppliers and will end up as mass-producer
in short order."
Wc aren't so sure, but \ve do feel this is a vital issue to whic
hardly anyone in the West has given much thought: Managem
of f.he macroeconomy may have a dramatic long-term elfeel
the f ~ n d a m e n t a lquality 01' the domestic production syste
Those public officials who have periodically felt it necessary
break the back of inflationav expectations by deflating the ec
omY may need to think anew about thc probable consequences
the production system. If nascent lean producers act to
themselves in a steep slump by jettisoning their most valua
asset-their people-the real cost of poor ~ e s t e r nmacroecono
managemcnt may prove even greater in the future than it
proved in the past.
More positively, widespread adoption of lean production ma
dampen both inflation and the business cycle. ~f mass produetio
is idealb suited to the survival of big companies through dee
cyclcs in demand, it may also be cycle-enhancing. ~ h is,~i t
penchant for massive inventaries, both of in-process parts an
finished units, would seem to exacerbate the cycle:
inflatio
prices. This move pushes prices up farther. Then, when the econ
OmY suddenly falters, the built-up stocks are worked off, deepen
ing the slump upstream in the production system.
Some observers have even ~vonderedif the lack of a cyclic
market in durable goods in Japan is a direct result
lea
production: An inventoryless, highly flexible system may sign$
cantly damp cyclicality.
The Japanese have another cycle damper in their arsenal i
Japanese companies receive a large part of their compensation
up to a third-in the form of bonuses directly tied to the profi
ability of the company. So when the market drops, a t least i
tlleor~,the company can dramatically slash prices due to low
operating costs and restore production to its former level.
In fact, this system has been tested only at companies
as
Mazda that have experienced a crisis independent of general
market conditions. The simple reason is that
have been no
CONFUSION ABOUT DlFFUSlON
deep c.eles i n the ~apaneseeconomy senerallv, so there has been
no real tesi of employee tolerante Sor large .rvage cuts."
WESTERN "CRREERS'VVESUS IRPANESE "GOMMUNITY"
~ h point
i ~ lea& directly to a second question confrontins the
future of lean production in the West. Why won't worlcers in
betier
companies temporarilY cutting wages s i m p l ~leave
opportunities in other companies or industries? The arlswer in
convention, practically al1 hiring in compaJapan is simple:
nies is from the I - > only
~ and,
~ ~ as ~noted,
~ compensation within
the company is based largely on seniorii~.Jumping ship lvould
be quite pointless, because the empl»yce would almost alwaYs be
worse off starting ar the bottom elsewhere ratlier than waiting for
the Situation to improve with thc current employer.
~ h situation
i ~ obviously doesn't prevail in the West. In addition, as we pointed out in Chapter 5, Western notions of careers
are quite different from the needs of lean production.
workers in the west place a very high value on having a
portable skill-something they can take with them if things don't
work out in a particular company. This concept is tied quite
tightly to western educational systems that stress discrete competences and certify students to prove thai the skills have been
~ t is very similar to the mind-set
attained, ~ h i ~ a b ~ skills
~
h
was-and
o
is-obsessed with mainof the skilled craftsman,
aithough
professional
workers in the
taining bis portable
~ e s rarely
t
see the parallel.
H
~ as we saw,
~
for ~the lean-production
~
~
system
~
to suc,
ceed, it needs dedicated generalists willing to learn manY skiiks
and apply them in a team setting. The problem, as we also noted
in chapter 5, is that brilliant team play qualifies workers for more
and better ,,lay on thc same team but makes it progressively
to leave, so a danger exists that employees who feel
trapped in lean organizations will hold back their knowlcdge or
actively sabotage the system. Western companies, if they are
to become lean, will need to think far more carefully about
personnel systems and career paths than we believe any have to
date.
252
In our discussion so lar, we've simply reported the obvious: Lean
production is diffusing rapidly in North America but mostly
under the leadership of Japanese companies. Yet, throughout
human history, foreign investment and ownership seem always to
operate on a razor's edge, continually tcsting the tolerante ofhost
countries. Recent shifts in American attitudes confirm this point.
initially the U.S. government was elated that the Japanese
would he plowing money into U.S. automotive plants. At the same
time, state governments fe11 over themselves with inducements to
invest in one state rather than another. And, as the investments
prew, the public came to accept the new planis as part of the
laridscape.
Recently, however, a new ione has emerged as the full logic of
Japanese investment has begun to dawn on public officials, the
U.S.-owned auto companies, the UAW, and U.S.-owned supplier
companies.
Fii-st, tlie Japanese companies do not believe they can turn
around existing mass-production facilities. Therefore al1 the
transplants are "greenfields" in car talk, or entirely new plants,
with the exception of NUMMI, which was a semi-greenfield, in
the sense that it had been permanently closed by General Motors
two yeais before it reoperied under Toyota management.
Second, the Japanese aren't building the transplants simply
Lo circumvent trade barriers or a temporarily strong yen. They've
discovered that iliey can make cars in North America as well as
they can in Japan and, even more important, that they can make
cars in North America better than two oC the three Americanowned companies. It therefore follows that the transplants will
keep on growing until the American companies improve their
performance and regain the initiative, or are eliminated.
Third, it appears that the UAW won't be able to organize
those transplants, whose owners have no l i n k to the U.S. companies. The union was badly defeated in an election at Nissan in
1989 and, so far, has been unable to gather the petitions it needs
for an election at Honda or Toyota. The Mazda, Diamond-Star
(Mitsubishi-Chrysler), and CAMI (GM-Suzuki) plants have been
organized, as has NUMMI, because, in each case, the plant has a
conneciion to a unionized U.S. firm. However, most observers
cxpect the plants of the Japanese Big Three (Toyota, Nissan, and
CONFUSION ABOUT DlFFUSlON
Honda) to grow the fastest. In consequence, the UAW has begun to
wonder about its institutional future if nonunion transplants
continue to displace the unionized plants of the U.S. Big Three.
I F it fails to organize plants, the UAW's only option inay be to
seek a limit on their expansion through political means. Recently,
the union has urged that vehicles assembled at the transplants be
subtracted from ihe number of cars each Japanese assembler is
allowed to import under the continuing Voluntary Restraint
Agreement. The logical effect of such a policy would be to set
permanent market-share limits on each Japanese company. This
would seem to guarantee the survival of the unionized U.S.-owned
companies.
Fourtli, as the Japanese companies rapidly increase their
domestic conient (that is, the Craction of a car's value manufact..~..ir.e.din the United States), U.S. suppliers are learning that it is
not easy io supply the transplants, for reasons we examined in
Chapter 6, and that it is hard to make much money doing so.
Putting al1 these facts together, we íind it hardly surprising
that members of the U.S. Congress, company executives, and
union leaders are beginning to wonder if the unarguable success
of the transplants is a cause for celebration or alarm. The North
American motor industry stands every prospect of being revitalized to regain world-class performance during the 1990s. The
massive trade deficit in motor vehicles is likely to shrink or even
disappear. However, this lean machine may be largely foreign
owned and nonunion if the American mass-producers don't improve their performance quickly.
We believe that the period through 1992 will prove the most
tense. If GM and Chrysler fail to go through a creative crisis, one
that breaks the logjam of old ideas and narrow interests and
opens the path to lean production, and if the economy should
slump badly during this period, we have great concerns about ihe
outcome. We've sho~vnjust how plans can go awry wlien we
discussed the European experience with mass production.
~
THE TRANSlTlON TO LEAN PRODUGTION I N EUROPE
But what of lean production in the current stronghold oí mass
production? As we have seen, ihe European auto industry is today,
after a fifty-year transition from craft production, ihe lcading
Proponeni of old-fashioned mass production-high volume, ion
product runs, infinitely fragmented work, "good enough,,
quality, enormous in~~entories,
massive
~ ~as we
d sa,
in the expel-ience of the young French factory rnailager, there h
been little pressui-e tlius ~ a tor achieve leanIiess, B~~ that press
will emerge as the 1990s unfold,
FirsL, tlie mai-kei may 1101 be as vigorous as it was in
1980s. Everyone can inake money in a sellerjs market. ~~d
opening of Easiern Europe inay produce a sustainable boom
the 1990s. Tlien again, gro~vingcongestion, environmental pro
lcms, a d approaching saiui-ario11i11 western E~~~~~may ho
delnand bel«w recen1 levels. Only a small slacking off in dema
\vil1 make a large diflerence in the profitability ofthe high.volum
E u r o ~ e a nfirms, who depend on very higli leve]s o~ capacit
utilization compared with the ~
~
~
~
Second, the Americans will be applying what theytve learn
in a
of desperate struggle in NortIl ~
~~~~d ~
already the most efhcient producer in E
~with the
~ except.
~
~
a few of iLs English opel-ations, such as Dagenham, tllat ne
really embraced mass production,
Third, and most important, the Japanese are arriving. I~ tl
1980s, Lhey were diverted to North America, a conlinent with no
local-content rules to delay the start-up of assembly plants and a
marltet where thcy liad already captured a 22.percent share
before the ercction o[ irade barriers.
In E u r o ~ e by
, contrast, market limits in France (3 percent
ItalY' (about 1 Percent), the United Kiilgdom (1 1 percent), an
SPain (a 40-percent tariffon imported cars) had held the
to an 11-percent share overall, ancl the rules ,-,fentry for
turing-in~ented quite amazingly by the c8free-enterprise
Thatcher government in Britain-included a requirement thdt
domestic content must reach 60 percent within two years of sLartUP and reach 80 percent a Sew years later."' ~h~~~restrictions, in
turn, have meant that Japanese companies can't just build assem.
Plants. They musi construct an engille plant and develop local
s"p~liersfor a host of components ~ i m u l t a n e o u s lgreatly
~,
raising
their start-up costs. Nevertheless, the Japanese are now rapidly
making their presence felt, as previously shown in ~i~~~~8.2.
Man~
in Europe have congratulated themsel~rcson their aggressive stance to~vardJapanese investment, ~h~~ view the U,S,
approach-thai is, practically free market access for any company
willin& Lo build an assernbly plant adding 25 percent or e\,en less
total value-as extremely mive. (cars made a l lhe
of each
JaPaU,S. transplants, that is, don't count against the quoca
nese finished.unit imports. And this holds true even if the assemler does nothing but screw together Japanese parts.) This aP,cscrewdriveroplants, they argue, with ver>'
roach yields
tile manufacturing value added. The heart of the industry, theY
rnaintain, will remain in Japan.
In our view, it is ilot a matter of being hardheaded or mive
ut rather f, understanding the interna] logic o[ lean ~
r
o
is a system based fundamentally on doing as much manufacturas possible at the Point of f ilal assembly. Once a lean producer
tarts down the path to assembly in a major regional market, the
ogic of the system ten& powerfully to bring the complete complement of production activities, including product development,
i
~ ~~d sooner
~
.
along
as well,
rather ~than later,~ as is happening
in
North America.
~ ~h~ i
&fecl
~ of the ~European. poíicies ¡S t« create a much
~
,
difficult
transition in the 1990s. As they begin the process of
adaptation, the European companies have sought to Perfeci
production as the Japanese have continued to perfect leanness,
farther behind than the Americans were in
and they are
198O4ff the pace by a factor of three or four in l n a n ~countries
in terms of fundamental productivit~.
~~~~don past experience, pressures will mount for a Fortress
E~~~~~ poticy-for example, permanent company market-share
limitations no matter where the product is produced. In fact, such
a policy was recently advocated by Peugeot president Jacques
Calvet, To do so would free7.e the current inefficiencies of the
and ensure that they slip even farther behind
E~~~~~~~
lean producers elsewhere, a disastrous outcome.
expect a compromise to be reached in m'hich lhe
However,
rate of J~~~~~~~advance is slowed but not halted by Policies from
Brussels, and in which the period of transition to lean production
is stretched out well into the next century. For example,
E~~~~~~~~~~~~~i~ ~ o m m u n i t y(EEC) will probably establish an
overall quota for Japanese imports to Europe and will establish
content requirements for assembly plants if their Europeanassembled vehieles are not to be subtracted from the import
quota, ~
~ ~ ~
Japanese
~ t investments
h ~ lwill~be ~
permitted
~ , in
any European country, and the quota will be relaxed over time.
sothe E~~~~~~~mass.production industry must evelltually learn
to compete with lean producers.
~
~
COMPLETING THE TRANSlTlON
257
transition to lean production can be complcted by the end ol this
century.
/ COMPLETING THE TRANSITION
0
It took more than fifty years for mass production
to spread across the world. Can lean production
spread faster? Clearly we think it is in everyone's
interest to introduce lean production everywhere
as soon as possible, ideally within this decade.
In North America, the full implementation of lean production
can eliminate the massive trade deficit in motor vehicles. We are
certain that when there is no longer any difference between the
North American and Japanese motor industries in productivity,
product quality, and responsiveness to changing market demand,
trade will more or less naturally come into balance.
In Europe, the home today of classic mass production, lean
production can quickly triple the productivity of the motorvehicle industry while providing more fulfilling jobs for factofy
workers, engineers, and middle managers. It can balance Europe's
motor-vehicle trade as well.
In many of the developing countries, lean production is a
means for rapidly developing world-class manufacturing skills
without massive capital investments. These countries would need
only to find markets for their new industrial capabilities, a point
we wiii return to in a moment.
Truly, the case for moving quickly is overwhelming. In this
concl~idingchapter, we offer some practica1 ideas on how the
.-.
The greatest obstacle in the path of a lean world is easy lo
identify: the resistance of the massive mass-production corporations that are left over from the previous era of woi-Id industry.
These companies-General Motors, Renault, Volkswageii, Fiatare so large and promineni in the industrial landscape of North
America and Western Europe that no government can allow them
to Eail suddenly. Yet many have provcd remarkably incapable of
reforming their ways in the 1980s.
What's more, most of the conventional means governments
use to aid their home-owned companies are counterproductive
over the long term. We've already noted the perverse effect of the
quotas negotiated by the U.S. and European governments a¡ the
beginning of this decade. While the quoras were useful in sending
a signal to the Japanese that foreign manufacturing would be
necessary in the long term (at a time when currencies were
sending just the opposite message in the short term), they generated massive profits for the Japanese companies to use in financing their direct-investment campaigns in North America and
Europe.
We think that it is neither practical nor desirable for these
massive Western enterprises to be swept aside by Japanese lean
producers, but they need more creative solutions than thc kinds
that have been conventionally proposed. These solutions must
take several forms:
First, every mass-producer needs a lean competitor located
right across the road. We have found again and again that middle
management and rank-and-file workers in a mass-procluction
company begin to change only when they see a concrete, nearby
example of lean production that can strip away al1 the cultural
and economic explanations of why the other manufacturer is
258
77% MACHINE JHAJ CHANGED JHE WORLD
be
succeeding. U.S., Canadian, and English mass-production plants
whole appl.oach to production is doomecl. So t h e ~seem
dying the death of a thousand cuts as the world's lean producers,
now al1 have a lean producer right across the road, but th
continental European countries have Iagged badly,
now inc]uding Ford, steadily rain t.i-ound.
relative
~h~ trick for investors and bankers when the crisis conles ¡S
performance on our productivity survey shows the result.
help-but only in return for the company's realistic pians
Even the Americans, Canadians, and English have only lis
to
for achieving \yorld-class performance by converiing to lean proexa*~les of lean factory practice to examine up clase. ~t is on[
duction, ~
~
~may also
~ nced
~ to help
~ out
~ by esiablishislg
~
~
n
now that lean research-and-development operations are bein
training programs for employces thai companics can no
established nearby. Thus the North Amei-icans and ~ ~ ~ha l i ~ paid
h
longer use, ~h~~~excess ,,torkei-s are the heart of the conversion
not made significant progress in adopting lean design, bu[
problem, ~ h ~because
t ' ~\vorkers at mass-production p1ants learn
exPect to see rapid improvement shortly. This trend is
a mass-production cornpanj7 c»llapses, most
occurring in supplier systems where many No& American
no skills, so
pliers are learning better methods from &ir work with
workers can q u a l i b only for the n ~ o s tenti-y-leve] jobs in other
displaced
transplants. They are then using this knowledge to improve th
sectors. ~
~
~
i workers~ cor meaningful
i
~ work ~
relationships with the U.S. assemblers. SO improvements, which
wiII be essential.
lean
a maior problem in coil\~crtingfrom mass
we'd normally expect lo come from the top d o ~ v n - a ~ ~ ~ ~ b ] ~ ~ ~
production is that in a highly competitive marketplace, where
working to make their suppliers better-are, instead, coming from
the bottom up.
share gl-owth is impractical, a substantial fs-action of the work
force is no longer needed. If the European volume producers wcre
Second, mass-producers in the West need a better system of
to lean productioi~today without gainillg fnaiket sharcs
industrial finance, one that demands they do better &ile supply.
to
they
would
need less than half their curreni work h c e . Doubtless,
ing the large sums that will be needed to turn these large companies around. Currently, most of the debate in the arca of finance
the market for cars and trucks would grow as competition pushed
focuses on low-cost funds for Western companies and ways Lo
prices down, but it is unsealistic to think that shedding *f labor
dis*antle the Japanese group system. ~ u c hproposals are no
can be avoided.
oneof the most important tcncts of the Toyota P1-oduction
doubt well intentioned but miss the point: Giving mass..producers
System is never to vary the work pace. Therefore, as efficiencies
more moneY to spend on inefficient product development, inefficient f a c t o operations,
~
and more sophisticated equipment than
are introduced in thc factor? or design shop, or as the rate
they
production falls, it is vital to rem«vc unneedcd ~vorkersfro* the
is bound only to make things worse in the long run. ~~d
system so rhat the same intensity of i.vorlc is maiiltaincd. Otherdismantling the Japanese group system would be to iunk the most
wise the challenge of continua1 improvement will he lost. Thc
dynamic and efficient system of industrial finance the world has
same is tme
mass-production companies convertin& lean
yet devised.
Tllird, most mass-producers will need a crisis, what we
production. E~~~~~ mrorl<ers must be removed completeb and
~as we
~ saw,
d had
,
have called a creative crisis, to truly change. ~
quickly from the production system if improvement efforts are
such a crisis in 1982. The company heeled
so far that the
not to falter.
Companies such as GM havc tried to do so by creating job
senior executives on the bridge were practicafly catapulted inro
worbers to be retrained for other iobs in the
the raging torrent. The result: A company previously plagued
banks o[
with i ~ ~ t e r ndissension
al
as executives tried to
own
company, ~h~ problem is that, realistically, there
nevcr be
any other jobs within the mass-production companies and lhe
careeSS and workers sougbt wages and benefits divorced from
ability of the companies to finance the .job banks will decay over
productivity gains suddenly had a new sense of purpose and team
sort of public support for .iob banks maY be
spirit in saving itself from oblivion. Organizational changes that
time as weil. so
had seemed impossible were suddenly easy. GM and the E~~~~~~~
necessary, and workers will have to be trained for posiLions
mass-producers, by contrast, have had periods of low profits and
outside the iradiiional manufacturing industries. This idea is
trises from 0 t h causes but have never had the sense that their
sternly resisted by government officials and union leaders in
t
THE MACHINE THATCHANGED THE WORLD
many Western countries, the former because of the up-front cost
to governments, the latter because the union loses strength
workers leave their companies. However, the alternative approa
of propping up mass-producers through trade and investment
barriers, so they can afford to continue their inefficient use o
human effort, is far more costly in the long term.
Obstacle 2: Outdated Thinking About the World Economi,
Once upon a time, not very long ago, most people thought that
the world economy advanced by moving the production of stmdardized, low-priced products-such as small automobiles and
tmcks-to new mass-production factories in neurly industrializing
countries. In the 1970s, the rise of Japan was often explained in
this wav.
Five years ago, when we started our project, many observers
expected that Japan would soon find it could no longer compete
in the export of small vans and trucks because of the strengthening of the yen and its e f f e ~ ton Japanese wages. Korea, Taiwan,
Thailand, and Malaysia, the next tier of countries with lo\v wages
and an educated, industrious work force, would collectively become the next Japan. These countries, it \vas argued, would
rapidly build up their economies the way Japan had-through the
export of small cars and trucks to the United States and Europe,
supplanting Japanese products in the process.
We never subscribed to this view, because we knew that lean
production is more than a match for low-wage mass production.
First, lean production dramatically raises the threshold of acceptable quality to a leve1 that mass production, particularly in lowwage countries, cannot easily match. Second, lean production
offers ever-expanding product variety and rapid responses to
changing consumer tastes, something low-wage mass production
finds hard to counter except through ever lower pr-ices. Continually dropping prices is unlikely to work, however, because a third
advaniage of lean production is that it drainatically lowers the
amount of high-wage effort needed to produce a product of a
given description, and it keeps reducing it through continuous
incremental improvement, as we saw in Chapter 4. Finally, lean
production can fully utilize automation in ways mass production
COMPLETING THE TRANSITION
261
cannot, further reducing the advantage of low wages. The expansion of the Korean motor industqr in the 1980s sums u p this
simation.
In 1979, Korea was no~vherein thc auto industry. Despite
government protection of the domestic marlret sincc 1962, the
Korean industry, consisting of four small producers, had not
gotten very far. Hyundai was the most advanced and, unlike its
rivals, was t o a considerable degree independent of the U.C., European, arid Japanese assemblers Its Pony model used an independently engineered body and an engine and transmission \\~11»11q~
manufactured in Korea from designs licensed from Mitsubishi. 11
had enjoyed some modest export success, mostly in third-world
markets, such as in Latin America, where ii sold on price. The
other small companies-Daewoo, Kia, and Done A-made vehicles solely for sale in the Korean domestic rnarket, using designs
licensed from European companies. Unlike Hyundai, they were
completely dependent on their European partners for technology.
The world economic crisis of 1979 and 1980 hit particularly
hard in Korea. Domestic sales collapsed. So did Hy~indai'sexports, once Japanese companies dropped their prices in order to
maintain their sales in export markets. This crisis gave the E;orean
Ministry of Industry the chance it had been waiting for-the
opportunity to rationalize the industry over the opposition of thc
chaebol (Korea's version o€ the Japanese Iceiretsu) in the way the
Ministry of International Trade and Industry (MITI) had wanied
to do i k ~ a p a nin the 1950s.
The Ministry p s h e d Kia and Dong A out of the i n d u s t r ~for
five years, while assigning Hyundai small cars and Daewoo larger
ones. Hyundai, in particular, took this directive as a sign to
proceed down the path to high-volume mass ~roduction.It began
to plan a new model, the Excel, which ivould he built in a massive
new factory in Ulsan, mostly for export to the United States and
Europe. The Excel was almost entirely based on licenses from
Mitsubishi in Japan. Indeed, it was practically indistinguishablc
in general specifications from Mitsubishi's Colt model. H ~ u n d a i ' s
strategy was simple: It would compete by underpricing the Japanese entry-leve1 cars, based on low wages and high volume.
For a brief period, the strategy worked brilliantly. Hyundai's
Excel arrived in the vital U.S. market in 1986, justas the Japanesc
were raising their prices to counter the strengthening of the yen.
Americans assumed that any Asian car, particularly one with a
Japanese design, would have Japanese quality. And, at a price
2lid
$1,000 below Japanese vchicles in the same size range, the Exc
seemed unheatable. Sales quickly grew to 350,000 per year, an
H ~ u n d a iscrambled to add capacity in K~~~~ with a
300,000-unit assembly plant.
The Korean Minislry of Industry \vas so impressed wit
Hyundai's triumph that it soon permitted ~i~ back into the aut
market. Kia \uould build a small car based on the ~~~d~ 12
be sold in the United States by ~ 0 r - das the ~
~Ir, addit
~
Daetvoo was allolh8edto build a second, sma]ler model, based
accounting for 4 pei-cent of the total market,
Then lhe K"rean strategy fe11 apart. Hyundai was, in fact, an
old-fashioned mass-producer, with low wages but a large number
hours expended per car. Wheii the Korean cuI-r-encybegar, to
strengthen r a p i d b against the dollar in 1988, and K~~~~~ auto
workers demanded large wage increases, a large part of the
Koreans' cost advantage disappeared. AL that point, the question
of quality emerged. The early Hyundaj cars sold in the united
had verY Peor quality, as shown in the J. D. power data we
used i* our assembly-plant studies in chapter 4,
In 1987, when the average Japanese car was reported by
consumers to have about 0.6 defects, the Hyundai cars had 3.1,
the
bcgan to spread, the Korean producers found it necessarY t" slash rices to sustain sales, jusi at the point when their
production costs wcre soaring. The consequence was that K~~~~~
in the Uniied States fe11 50 percent betweerl 1988 and ,ggO,
The ncxl Japan was no longer the next j a p a n , l
What's more, by the late 1980s, it had become apparent that
there would be no next Japan, even if a developing country created
a lean-production industry that could match the product quality
and labor productiviry of the best lean producerc, japanrs
success
had so sensitized the world trading system lo massive inflows of
industrial
products fi%m one region to
that no country
could realisticall~hope to pickup where Japan left off, lndeed, at
the peak ~f the Korean auto sales in North ~~~~i~~ in 1988, the
U.S. government pushed the Korean government hard to reduce
its growin& overall trade surplus by 50 percent, which the K ~ ~
ans did.
Hyundai became convinced that lo protect its access ío rhe
North ~~~~i~~~ market, it would need to follow the Japanese
Its
transplants in building an assembly plant in North
100,OOO-unitBromont, Quebec, plan1 opened in 1989 t" assemble
the mid.size sonata, a new model it hoped would restore its
rtunes in ti,e ~
~ American
~
t market.
h
The idea of a con'pany
om a developing country building a majar manufacturing [aci'in a highly developed, high-wage country wo"ld
nthinkable
i
~ only ~five years
. earlier, a t a time when most
inexorable
drift of lo\v-iech manufacturin~,inciudredicted
cheap autos, out of the developed world. After all, Korea's
ire advantage was expected to be low wages. But because
Korean wages are rapidly con\~ergingwith those in Nwth Alnerica, and political conciderations require manufacture within the
Korea is now assembling cars in Cariada.
North American
~
~
What do we conelude from the experience of I3!íundai ancl the
other Korean producers? Thar the world economu, in a short
period, has =hanged in remarkable ways. First, the triurnph oí
quality
lean production has ereated a new threshold for
that no producer can hope to offset merelv through low prices
of
a result, producers in the nexL
low wages.
developing countrjes must become lean producers as weli. As we
will see shortly, it is quite feasible for them to move toward their
al in the 1990s.
secona, those developing countries mastering. lean production will need to tilink anew about the market for lhcir
products, parlly,
they ~ h ~ ulook
l d a t heme, because the productivity gains of leal, production should bring motor \'chicles jnio
a much larger fraction ol domestic consumers. In Brazil,
for example, we found that rifty hours were needed t" a*sernbie
our standard small car co,npa~edivith thivree~zat the besiJapanese
lean producer, ~~t surprisingly, the Bra~iliancar markct has
1 mil]ion units for many Years. About Onebeen stuck at
the difference in performance is due to greater automathird
tion in the highect-productivity Japanese plant, but
introduction of lean production without advanced technolo~y
should cut the leve] o[ effort in Brazil in half, opening a vast new
domestic market.
The developing countries should also look for regional mar.
kets. Indeed, ihe most striking feature of the worlcl economy in
the past íew years is the sudden reorienlation of the trading
264
patterns in manufactured goods from cross-regional, across the
great oceans, to intraregional, within the great regions-North
America, Europe, East Asia.
The motor industry is perhaps the leader in this trend. Exports from Japan to Europe are stable, while Japanese and European exports to North America are failing dramatically and European exports to Japan are growing dramatically from a veqr
small base. (We don't, of course, count as exports cars manufactured in the United States and Europe by Japanese companies.)
What we expect by the end of the decade is a much lower volume
of total exports between regions, a greater balance in the remaining trade flows, and a focus for the remaining cross-regional trade
on niche-type, specialty products. This approach, of course, is
preciseiy the one proposed for the post-national lean enterprises
we described in Chapter 8.
Meanwhile, within the great regions, the flow of products
among countries should increase dramatically. Let's begin with
North America. The United States and Canada began to integrate
their auto industries in 1965, when the Canada-U.S. Auto Pact
went into effect. For the participating assembler companies, the
U.S. Big Three, this meant that cars and trucks could be made in
one country and shipped to the other for sale without paying
tariffs-as long as the assemblers met modest Canadian requirements to Iteep Canadian production roughly proportional to Canadian sales. (This point quickly became moot, as Canada began
to run a substantial trade surplus with the United States.) In
1989, the Canada-U.S. Free Trade Agreement set in motion the
final process of integration of the automotive market by eliminating by the mid-1990s al1 remaining tariffs on thc flow of parts
between the two countries.
An interesting issue in [he North American region is Mexico.
For a period of thirty years, from the beginning of the 1960s,
Mexico tried to develop a domestic motor industry that could
supply the Mexican market with al1 its needs. To achieve this goal,
the Mexican government in 1962 prohibited the imports of finished vehicles and imposed high local-content requirements on
the five foreign companies-Ford, GM, Chrysler, Nissan, and
Volkswagen-building cars in Mexico.'
The policy was both a spectacular success and a spectacular
tailure. By 1980, Mexico had a 500,000-unit motor-vehicle industry producing vehicles with perhaps 50-percent local content.
COMPLETING THE TRhNslTloN
Unfortunately, the industsy-with market-share restrictions and
a host of other protections for the domestic assemblers and parts
suppliers-was totally uncompetitive in both cost and quality in
the world market. With five producers making three or four
separate models each in a 500,000-unit market, average annual
production totaled about 25,000 units of each product, far too low
even for today's lean producers to make economically. What's
more, the Mexican plants were in no way lean. Even Nissan, a
lean producer in Japan, employed a combination of craft and
mas-production methods at its Cuernavaca plant.
Mexican policy might have continued on its course except for
the economic collapse beginning in 1981. As domestic demand
tumbled and Mexico's foreigu debt mounted in 1983, the government examined its auto policy. Its initial strategy was to push
even farther down the path to mass production by limiting each
assembler to a single product while raising tlie required domestic
content level. Government officials reasoned that although Mexican car and tmck buyers would have vesy limited choices, economies of scale should cause the cost of cars and tmcks to fall as
volume increased. What was more, the Mexican trade deficit in
automotive products should fa11 as domestic content grew.
It soon become apparent that this strategy wouldn't work.
The domestic masket was simply too small and the protected
domestic producers too inefficient. Mexico would need to join the
world. The first step was to permit Ford to build a new assembly
plant in the northern city of Hermosillo. This plant had no
domestic content requirement as long as it exported the great
majority of its output.
The Hermosillo plant also provided the first opportunity to
experiment with lean production in Mexico. At this greenfield site,
Ford applied what it had learned from Mazda in building a
Mazda-designed car, sold in the United States as the Mercury
Tracer. Hermosillo was a great success in terms of productivity
and quality. Mexican workers embraced lean production with the
same speed as American workers at the Japanese trans~lantsin
North America and at Ford's own U.S. and Canadian plants.
Hosvever, the plant failed to meet its cost targets, because it was
assembling its cars entirely from parts shipped from Japan. A s
the yen strengthened, Hermosillo, a plant envisioned by Mazda
and Ford in the early 1980s as a way around the U.S.-instigated
quota on Japanese finished-unit imports, suddenly made no sense.
What does make sense-and is also consistent with the concept of
268
JHE MACHINE THAJCHANGED THE WORLD
North America and Europe in its development. Only a few years
ago, the individual economies of Japan, Korea, arid Taiwan w
struggling to boost their exports of finished manufactured goo
for the North American and European markets. They seeme
almost oblivious of each other and were highly resistant to accep
ing manufactured goods imported from their neighbors. T
situation is now changing rapidly, due partly to trade barriers
and currency shifts that close off markets in the other regions,
partly in response to the movement toward regionalization in
Europe and North America. In 1989, for the iirst time since World
War 11, trade within tlie East Asian region excceded extraregional
trade with North America and Europe.
The logic of motor-vehicle industry development in East Asia
is similar, save in one respect, to that of North America and
Europe. We expect more basic vehicles to be made in top-tobottom manufacturing complexes in the developing countries of
the region for sale in a11 countries of the region. We also expect
the manufacture of more complex and expensive vehicles to be
focused in Japan for export to other markets in the rcgion. Indeed,
this trend is already beginning. Hyundai, Kia, and Daewoo al1
plan to begin selling entry-leve1 vehicles in Japan in 1991. The
Korean domestic market, thus lar closed to finished Japanese
vehicles, will be opened a crack at the same time. While the
domestic Japanese industry, unlike those in Western Europe and
North America, is unlikely io prove better off under this arrangement, it can easily be no worse offif its exports of more luxurious
vehicles grow enough to offset decreased domestic production of
basic vehicles.
The anomaly in East Asia, of course, is China. Until spring
1989, it seemed to be moving toward a more open stance regarding its economy and the world and might logically bave entered
into a regional East Asia market on at least a limited basis.
Perhaps it can yet do so in the 1990s, but for the moment the
Chinese industry is still focused inward, pursuing a combination
of extremely rigid mass production in its two volume-production
complexes in Changchun (No. 1 Auto Works) and Hubai (No. 2
Auto Works) and inefficient low-quality craft production in about
a hundred additional vehicle-manufacturing facilities spread
throughout
---...
- China.6
This disastrous combination givcs China the distinction of
having the world's largest motor-vehicle industry in terms of
employment (more than 1.6 million workers) and one of the
COMPLETING JHE JRANSIJION
269
smallest in terms of output (a projected 600,000 units in 1990).By
contrast, in 1989, 500,000 employees in the Japanese auto industry produced 13 million vehicles, indicating a productivity gap 01
about seventy to one between two countries separated by a hundred miles across the Sea of Japan.
So much lor the three great regions comprising about 90
percent of the current-dajr motor-vehicle market. What o£ those
countries, such as Brazil and Australia, with substantial motor
industries, and others such as India, with substantial as~irations?
Where do they fit into the emerging world of regions and regional
production systems? Our belief is that they must look primarily
to their own regions for markets, but in creative ways. Let's take
the two very dilferent examples of Brazil and Australia.
Brazil set out in the late 1950s to build a top-to-bottom
motor-vehicle production system It ~ e r m i t t e dthe multinational
car companies, notably GM, Ford, Volkswagen, and Fiat, to own
100 percent of the equity in their Brazilian operations but insisted
that they quickly convert from kit building, using i m ~ o r t e dparts,
to the use of practically 100-percent Brazilian parts in each
vehicle. By the mid-1960s, in the midst of the Brazilian economic
miracle, this goal had heen achieved. The Brazilian industry
reached 1 million units of production a n n ~ a l l y . ~
Unfortunately, for twenty years now, Brazil has been a story
of stagnation. As we noted, the mass-production complexes built
in Brazil were a notable achievement compared with the alternative-complete dependence on irnports. However, these plants
now lag far behind the world pace in terms of productivity and
product quality. In addition, in the early 1970s, after oii prices
soared, the government required that the industry introduce alcohol-fueled engines, a requirement that focused the industry's
product-development energies on a technology that has found no
market elsewhere in the world. Meanwhile, the number of years
each model \vas kept in production soared to fourteen years in
Brazil, nearly four times the Japanese standard.
For a brief period in the mid-1980s, the Brazilian industry
thought it had found a new strategy: It would take advantage of
its low wages by exporting cheap cars to Europe and the United
States. (The models in question were the Volkswagen Fox subcompact sold in the United States and the Fiai Duna sold ir1 Euro~c.)
This was a Latin variant of tbe Korean strategy, which met with
270
211
the
pattern of high hopes, based on initial sales, followed b
despair as currencies shifted and product shortcomings offset a
initial price advantage. Fox sales in the U.S. market, for example
fe'' from a peak of 60.000 in 1987 to 40,000 in 1989, GM, mean
while, cancefed a tentative plan for Brazilian production of
mini-mini van, based on its German opel ~
~ mainly
d fo ~
exPort to the United States,
A more Promising path for Brazil in the 1 9 9 0 ~will consist o
three elements. First, lean producers must show ~
~the wa
~
~
toWard world-class manufacturing. Honda's motorcycle plant at
Manaus,
uP the Amazon, has clearly demonstrated that lea
production can work in Brazil under the most demanding cond tions, but automotive exampies in the Brazilian industrial heartland near Sao Paolo are e s ~ e n t i a lIntroduction
.~
of lean produccan dramatically reduce production costs to spur the
domestic market, where only the upper-middle
can
"OW afford the OutPut of the inefficient nlass.production auto
industry.
Second. Brazil needs to open its industq to imports of whole
vehieles ami Parts, SO that real cornpetition &"ill be introduced
into what is now a tight oligopoly. Because ~
~can hardly
~
~
i
afford t" run a trade deficit in motor vehicles, given its massive
foreign debt, it will no doubt need to require that producers
their trade. However, a truly competitive market can still
be achieved with a flexible policy. The new ~~~i~~~
I)ecree
shows one way to accomplish this
Third, Brazil wifl need to integrate its production system
with jis neighbors, beginning with Argentina,7 B~ setting in moa regionalization process as it brings production costs down,
Brazil can de velo^ a massive Latin American growth market that
will not d e ~ e n don favorable trade policies and currencies in the
other great
regions. While productive trade with those regions
should
be possible, ir will not be the key to ¡he strategy. Thus
Brazil and its neighbors can control their own fate,
c.ass
Presents perhaps the most diffificult instante
a
counts.
with
a small and highly developed motor.vehicle industs.
but with
insufficient dornestic market and, thus far, a jack of
regional outlook. The Australian government decided in the 1960s
that it would develop a top-to-bottom auto industry to replace
both i m ~ o r t e dvehjcfes and the kit assembly ofparts produced in
Europe and North America. By the end of the 1 9 6 0 ~i t liad done
so but witll all the disadvantages of mass production in a Iowvolume, higllly protected market. Despite Australia's elforts i n the
1980s to consolidate the five producers into thrce more
production syslems and the presence of several JaPane" producers, our I M ~ assembly.p~anl
p
surveys halie found productivi'y
~ quality
~
,
and
levels
lar off the standard set by lean pr0ducers in
fapan and North America.
For a ivhile in the mid.1980s, Australia thought that perhaps
i
l
could
succeed by following Korea's example. Ford proPosed
it
a ~ a z d 323
a modibed as a
to export a specialty vehiele,
in
the
United
States under lhe name
ble roadster, to be
a time when the Australian dollar was
Mercurv Cap,.i, ~ hwas
i ~
very w;ak and the ~ ~ ~ ~dollar
i c very
a n strong. HavciJer, by l h c
for production and a numher of qLiality
time the car was
problems had been resolved, currencies had shifted and the
no longer made m u c economic
~
sense? Thc effort illustrates Once
extraregional
export strategy in a world o'
more the risk of
fluctuating currencies.
The logical path for ~
~ ~ o utl be
d ~to reorient
~
l its ii11dustry
~
toward the oceanic regional market including Indonesia, Singapare,
l
and the Philippines. ~~~h country within lhis regioti mi&t
balance its motor.vehicle irade, but, collectiilcly, by permitting
gain the
cross-shiprnent ,f finished units and parts, t h e ~
seale needed to reduce costs and let lean prodllction flourish.
Australia, as the rnost 'dvanced country in the region, presumahly
produciion on coillplex luxur)' "ehiwould concentrate its
cheap,
eles, while Indonesia at the other extreme, would
entry-leve1 products.
Unfortunately, nothing of this sor1 has happened.
views itself as part of the developed ivorld and thinks naturally
exporting to North ~
~Europe, ~and even ~Japan, while
i
~
nesia thinks of itself as part of the developing world o' lhe
Associalion of south ~ ~Asian
s Nations
t
(ASEAN) countrics and
focuses on developing trade \vith Malaysia, the Philippines, and
by poolThailand. Repea& efforts to develop an
ing parts produced by difrerent companies in each countrY have
come to nothing, because it doesn't make sensc to do so in
the multinational asseml->lcr
of the commercial sirategy
cornponents firms.
Thus the oceanic countries o[ the Southern Hemisphere constitule a region still u,aiting io happen. The same can be said of
countries of the lndian subcontinent and those of southern
~
272
THE MACHINE THAT CHANGED THE WORLD
As the rest of the world pursues a path of regionalization
in the 1990s, we expect that regional thir~kingwill grow in these
a-eaS as well. The combination of regional seale and lean production can he a particularly powerful stimulus to g r o w t ~if the right
policies are followed.
Ohstacle 3: lnwaid Focus 0f the lapanese Lean Pro#ucers
The
obstacle
lean
a lean world is, in fact, the japanese
Producers themselves. How can this be?
of you will, no
doubt, have concluded that we think everything these companies
do is good, compared with the bad practices ofthe westernmassproducers. In olle way this impression is accurate: ~h~~~companies have provided an invaluable gift to the world by pioneering
a new waY of making things that really is superior. ~~t in another
waY they laclc a final and essential innovation: the ability to think
and act globally rather than from a narrow national perspective.
Anyone who reads the newspapers is aware of the growing
backlash to Japanese direct investment in ~
~ ~~~~i~~
~
t and h
EuroPe, what the Japanese themselves cal1 investment friction,
We regard this trend as a much greater thl-eat to the eventual
creation of a lean world than trade barriers on finished
and
parts ~ o u l dever be. This is because, in the worst case, it can lead
t" investment barriers that permanently seal off Norrh ~
~
Europe,
the other regions from Japanese lean competitors
that can force everyone to become lean,
W h ¡S~ this backlash building when the J~~~~~~~
companies
are creatin& new jobs in new manufacturing complexes making
cars, trucks. and parts at levels of quality and productivity equal
t"
of the heme plants in Japan? Partly it stems from the
lhreat these facilities represent to established institutions-massproduction companies and mass-production unions, l+iction, for
these reasons. is an inescapable component of change and
progress.
However, there is another, more fundamental reason for frie.
M a n ~government officials, managers, and workers in the
West perceive that the Japanese lean producers are offering two
'lasses of citizenship in their o r g a n i ~ a t for
i ~japanese
~ ~ ~ ~ ~
workers, a second for foreigners; one for J~~~~~~~suppliers,
another for foreign suppiiers; and one for hpanese
group mem.
bers of their keirersL1, but none at al1 for foreign companies. As
Westerners watch the seemingly inexorable advance of the Japanese companies, this second-class citizenship begins to seern unacceptable, one manager at GM remarked, "1 can hoPc
get
GM,
but
1
can
never
hope
to
rise
above
the
to the top
japanese
foreign suhsidiaries, no matter how
of one of
superior my performance.u Growing investment friction is the
result. The outcome is uncertain.
Executives at the Japanese companies are acutely aware
don, now
this problem and have given it much thought. One
pursued by several of the auto coinpanies, is to a p ~ o i n t
native managers to head their manufacturing op"ations i n
and E
~similarly,
~
a number
~
~of Japanese
~
companies
.
are designating native supplier companies as their source for
certain categories of components Governments in both regions
are supporting this approach through restrictions on visas for
Japanese employees at the new facilities and, in EuroPe, lhrough
strong pressures to attain high levels of domestic content as 'Oon
as possible, ( ~ latter
h ~policy raises the cost and launch time of
the initial facilities substantially unless most parts are ohtained
from existing domestic suppliers.)
The consequence, we fear, will be a repetirion of F ~ r dexPe'~
domesriente in ~
~after 191
i 5 . The
~ wholesale
~
substitution
i
~
tic managers and suppliers, to deal with investment friction,
quickly degraded the performance of Ford's production Wtem
~ toward ~the existing
i
~
~ While Ford
,
English
level.
did sPur
producers to adopt new w a y ~the
, full benefits of mass production
were never achieved.
In Our
This is not an idle fear, based on events long
of the transplants in North America and
assembly..plant
Europe, we found strong evidence that those plants that perrorm
best are those with a very strong Japanese management Presence
in the early years of operations, and those that have moved skowly
and methodically to build up their domestic supply base. The
performance of other plants, which have turned over most mansenior
agement to ~
~ Americans
~
t and
h Europeans recmited
levels from western car makers, and who have hurried1y
bled a supply team, is better than the Western average but in
many cases not as good as the Western c o m p a n ~ - F o ~ ~has
-~~~~
taken lean production to heart.
that the "~apaneseness"o[ the management
It should be
and the suppliers is not the issue. Rather, it is how well transplant
THE MACHINE THAT CHANGED THE WORLD
managers and suppliers understand lean production and ho
deeply they are committed to making it work. Unfortunately, a
the moment, a large fraction of the world's managers with knowl
edge and comrnitment to leanness are Japanese.
We would suggest that a better approach for the Japanes
companies will he to huild a truly global personnel system i
which new workers from North America, Europe, and every othe
region where a company has design, engineering, and productio
facilities, are hircd in at an early age and given the skills, includ
ing language skills and exposure lo rnanagenierit in differen
regions, needed to become full citizens of these companies. This
will mean an equal opportunity to head the company someday.
Similarly, the Japaiiese lean assemblers will need to form
supplier groups in each region where they operate by exchanging
shares in supplier firms and offering Full citizenship. They will
also need to regionalize their equity base and borrowings so that
shifting currencies will not hinder the most appropriate deployment of production in each region. Finally, a tmly important
advance in terms of its visibility will be Sor the keiretsu to include
foreign companies in their membership. For example, those keiretstc, such as that OS the Dai-Ichi Kangyo Bank, with weaker auto
companies among their members (Isuzu and Suzuki in this instance) might invite a strong Western car company to join. On the
other hand, the one Japanese car company unaffiliated with a
keiretsu, Honda, might wish to form an international keiretsu
consisting of Western manufacturing companies and a bank.
For any of these innovations to work, a clear two-way understanding will be essential. Western companies and employees will
need to embrace the concept of reciprocal obligation, making a
long-term commitment to rhe company o r the group. Japanese
companies, in turn, will need to abandon their narrow national
perspective and quickly learn to treat foreigners who accept the
obligations involved as full citizens.
We are well aware how difficult these innovations will be to
implement. American and European companies have struggled,
in many cases for decades, to provide full citizenship for foreigners in their orgaiiizations. Et there are ski11 no foreigners in the
senior management of General Motors or on its board, and it
made headlines recently wheri Volkswagen appointed Daniel
Goudevert, a Frenchman, as the first foreigner on its management
board.
In addition, the Japanese will need to address issues of ethnic
COMPLETINGTHE TRANSlTlON
background and gender in providing citizenship Sor foreign employees, issues that they have not confronted in Japan (where
there are practically no minorities and where women are notably
absent Srom senior management) and where Japanese practice is
far behind Western norms.
Nevertheless, the Japanese must logically be the innovators
in devising postnational, multiregional corporate forms and providing full citizenship to their employees and suppliers drawn
Erom many countries and regions across the globe. They have the
financia1 resources that many Western companies lack and they
have the need; they risk investment barriers and other impediments to the expansion of their production systems if they don't.
They should start right away by declaring their intention to
proceed as postnationals (companies where nationality is not
related to promotion prospects) and by implementing postnational personnel, supplier, finance, and keiretsu systems that the
outside world can examine.
"Transparency"-the
ability for outsiders to see the system
in action, understand its logic, and verify its performance-is
critica1 to Western acceptance, because of the long time lag
between initiating such a system and proving that it really works
(Sor example, as young employees entering a t the bottom rise to
the top). One highly visible way to demonstrate their intent would
be for Japanese companies to assign newly hired Westerners to
work for several years in Japan, where there are ~resentlypractically no permanent non-Japanese employees of the big companies.
Only a p b l i c and emphatic commitment to these final organizational innovations-which Western firms must match as wellwill ensure the triumph of lean production, for the Japanese
companies and for the whole world. Such a commitment will also
provide part of the essential glue to hold together the emerging
world regions in North America, Europe, and Asia-regions no
I----a-n n ~ e united
r
by the familiar EastiWest conflict and in danger of
drifting apart in the twenty-first century
In Europe, the idea of mass production was a problem not
just for the auto industry, but in every industry. On one level, the
intellectuals, particularly on the left, embraced the idea of mass
production as the obvious means to elevate the living conditions
of the masses. Soon the images o[ mass production and modernity
were a central theme of European art. However, back at the
lactory, in every type of manufacture, the poor fit between the
requirements of mass production and the craft orientation oE both
workers and managers insured that adoption of the new techniques was very slow. The lack of an integrated European market
was a Curther impediment. It was only after World War II that
mass production was fully embraced across the industrial landscape of Europe, in many cases through use of "guest workers"
from other countries and cultures who were willing lo tolerate
the monotony of'classic mass production in the factory.
When Henry Ford and Alfred Sloan created mass
production, the ideas they incorporated were in
the air al1 about them. Everywl~erethere was a
sense that the older craft-based modes of production had reached their limits. What was more, many parts of the
mass-production system had been tested previously in other industries. The meat packing industry, Sor example, had pioneered
moving "disassembly" lines for cutting up carcasses before the
turn of the century. In the 1890s, the bicycle industry had pioneered many of the steel stamping techniques and dedicated
machine tools Ford later used. Even earlier, the transcontinental
railroads had developed many of the organizational mechanisms
for managing large firms operating over vast areas.
But Ford and Sloan were the first to perfect the entire system-plant operations, supplier coordination, management of the
entire enterprise-and to couple it with a new conception of the
market and a new distribution system. Thus, the auto industry
became the global symbol of mass production.
The complete system spread rapidly to other industries in the
United States in the 1920s and was soon embraced by practically
al1 volume manufacturing iridustries. In addition, mass production was tried, without much success, in one-of-a-kind craft industries-in particular housing, where a number of entrepreneurs set
out to become the Henry Ford of the home.
Ilfi
Just as Ford and Sloan were swimming in the sea of new
ideas, the postwar chaos in Japan created a fertile environment
Eor new thinking. Many of the techniques Eiji Toyoda and Taiiclii
Ohno built into their lean production system were being tried a t
the same time in other industries. For example, the qualityenhancing ideas of the American consultant W. Edwards Deming
were adopted at about the same time by many Japanese companies across a range of industries. A number of other ideas were
forced on these inventors by larger social forces in society, in
particular the need to treat workers as fixed costs once it became
apparent ~ h a hire-and-fire
t
labor policies would be strenuously
resisted by employees.
However, like Ford and Sloan, their achievement !ay in putting al1 the pieces together to create the complete system of lean
production, extending from product planning through al1 the
steps of manufacture and supply system coordination on to
the customer. Thus, the auto industry once more changed the
xvorld and has become the global symbol of the new era of lean
production.
What's more, as we have seen, lean production combines the
hrst features of both craft ~roductionand mass production-the
ability to reduce costs per unit and dramatically improve quality
while at the same time providing an ever wider range of products
and ever more challenging work. The final limits of the system are
not yet known and iis diffusion, both within the auto industry
and to other industries, is still at an early state-about where
~
278
mass p r o d u c t i o n was i n t h e e a r l y 1920s. Yet i n t h e end, we believe,
l e a n p r o d u c t i o n w i l l s u p p l a n t b o t h mass p r o d u c t i o n and the
r e m a i n i n g o u t p o s t s o f c r a f t p r o d u c t i o n i n al1 a r e a s of i n d u s t r i a l
e n d e a v o r t o b e c o m e t h e standard global p r o d u c t i o n s y s t e m of t h e
twenty-first c e n t u r y . T h a t world w i l l be a v e r y d i f f e r e n t , and a
much b e t t e r , p l a c e .
End Notes
1. Alan Altshulcr, Martin Anderson, Daniel Joues, Daniel Roos, and Jamcs
Woinack, Tlie Fnic~reofthe Auioinohile, Cambridge: MIT Prcss, 1984.
CHRPTER 1
1. Peter Drucker, The Coricepr of rhe Coy>orutio!.i, New York: John D q , 1946.
2. Sce, fur exarnple, Ford Motor Cornpany Chairman I-iarold Poling's specch
to the Automolii>eNetv.s World Congrcss, January 7 , 1990, in which he
estiriiated that worldwide "rxcess capacity" in Lhe motor vehicie indusiry would rcach 8.4 rnillion units in 1990.
3. For an exccllent assessmcnt os General Motors' predicamciii, see Maryann Keller, Rude ilciiukening: Tiie Rise, Fui1 urrd Struggle for Recoi>cn'iii
Cerieuul blotors, New York: William Morrow. 1989.
CHAPTER 2
1. The material in this scction on Evelyii Ellis and his car was obtained in
ihe archives o [ the Science Museum, London. It c«nsists os newspaper
accounts os Ellis's exploits and an interna1 background memo prepared
by the museum staff on the 1894 Panhard car bclonging tu the museum.
2. The material on Panhard et Levassor is Srom James Laux, In Fivst Csur:
The Freizch Auto Indrsl- lo 1911, Liverpool: Liivrpool University Prcss,
1976.
3. Fnrd acquired rnajority control of Aston Martin in 1987.It alsu acquircd
the small British sports car builder AC in that year. Other craii produccrs acquircd by rnultinational auto Firrns in the 1980s wcre Lotus
(General Motorsj, Fcrrari (Fiat), and Lamborghini (Chrysler).
END NOTES
4. Ford proposed this term in his 1926 article for the Encyclopedia Britannica, "Mass Production" (13th edition, Suppl. Vol. 2, pp. 821-823). Many
others a t the time called his techniques "Fordism."
5. Two extraordinarily useful studics of mass production in the Cactory, as
pioneered by Ford, are David Hounshell, Fronz the Americarz Systern ro
A4ass Prodzicrion, 1800-1932, Baltimore: ~ o h n Hopkins
s
University Press,
1984, particularly chapters 6 and 7, and Waynr Lewchuk, ilnzeri
Echno1og.y urzd ¡he Britisl? Ehicle Indz~strq:Camhhdg-e: Cambridge U
versity Press, 1987, particularly Chapter 3. The acrount given here of the
origins of thc Ford systeni is taken from thesc sourccs, unli:ss otherwise
indicated.
6. In 1919 the eiitirc vchicle assenihly department at Highland Park e
ploycd only $3,490 of capital cquipment (Lewchuk, Arnericnn E c h ~ i o
n. 491
- ,.
7. Ford's ability tu cui prices during the life of the Model T is summarized
by William Ahernatliy, Tlze Prod~~crii~iry
L)ilei~zrna:Roadbluck tu Ir~tzoiarion in the Autonzohile Industiy, Baltimore: Johns Hopkins University
Press, 1978, p. 33.
8. The Ford hiant~al,Detroit: Ford Motor Company (no date), pp. 13, 14.
9. This survey is citcd in Danicl Raff, "Wage Delcrmination Theory and the
Five-Dollar Day a t Ford," Ph.D. dissertation, Massachusctts Institute of
Technology, 1987. an interesting study of the social implications of Fordrs
.
<vztem
-2
10. ~ h i sovcrsight lnay help explain why the productivity of the wholc
factory did not improve a t the same rate as that of the assembly lines.
See Lewchuk, pp. 49-50.
11. Alfred D. Chandler, The Visible Hand: The M a n u g e ~ a lRevoluiion in
Anzerican Business, Cambridge: Harvard University Press, 1977.
12. This information and subseqnent material un Ford's organizarion and
operations are taken from A11an Nevins and Frank Ernest Hill, Ford: The
Tirrzes, the Man, rhe Company, New York: Scribners, 1954; Allan Nevins
and Frank Ernest Hill, Ford: Expansior~and C h a l l e n ~ 1915-1932,
,
New
York: Scribner's, 1957; and Mira Wilkens and Frank Ernest Hill, American Enrerprise Abr.oad: Fot-d un SLT Coniinents, Detroit: Wayne State
University Press, 1964. The specific information un U.S. assembly plants
is from Nevins and Hill, Ford: Expansion and Chullenge, p. 256. The figure
for foreign assembly plants is derived from Wilkens and 1-Iill, Appenrliv
7
..
13. Alfred P. Sloan, My Years with General Morors, Garden City, Ncw York:
Doubleday, 1963. Peter Drucker prcsented his own codification in The
Concept of the Corporation in 1946. Henry Ford 11 read this voluine when
taking over from his grandfather that year and set out to remake Ford in
CM'S image.
14. For the best explanalion of the logic of mass-production unionism see
Harry Katz, Shifiitig Gears: Changing Labor Relations in the U.S. Automobile Indusrry, Cambridge: MIT Press, 1985.
1. lbyotu: A History of [he Firsr 50 Years, Toyota City: Toyota Motor Corpo.
ration, 1988, provides a useful summary of Toyota's history.
281
2. The Toyota production total has been calculatcd from -'oía:
A Nisio?.
p. 491. Toyota had also produced 129,584 trucks between 1935 and 1950,
mostly for rnilitary usc. Thc Roure production figure includes 700
vehieles asscmhled a t thc Kouge and 6,300 kits of parts Ford shippcd tu
its linal assemhly plants spread across the United Statcs.
3. R,)>ora: Tlie [;irsi 30 Years, Tokyo: Toyota Motor Company, 19b7. pp. 325.328 (in Japancse).
4. In <he intercst ol' hrei,ity ivc have slcippcd over tlie many conceptual
contributions of the ~ o i o t aMotor Curporaiion's foiinding gcnius, Kiichiro Tuyoda. Kiichiro Toyoda had a tirimher of brilliant insiglits in the
1931Js, inspired in part by his own visit tu Ford in Detroit in 1929. These
inciudcd thc just-in-time supply coordination system. 1-Iowc\~er,tlic chaotic conditions in Jüpan in thc 1930s prevcnted him from implcmcnting
niost of his idcas.
j. For an excellcnt summary of the devclopment of thc Tuvota MotoiCorporation and the tcchniques of lean production. see Michael Cusumano, The lapa~zeseA~rtorrrobileInrlusiiy: Tech~lolog?und hia~zugen~eilt
ar
Ni.ssan and Tuyotu, Camhridge: Harvard Uni\rcrsit). Press, 1985.
6 . Toshihiro Nishiguchi, "Strategic Dualism: An Alternativc in Industrial
Societies," Ph.D. dissei.tation, Nuflield Collcge, Oxfor<l llnii~crsity.1989,
pp. 87-90, pro\,idcs a good analysis »f the consequcnces o[ tlie ncw labor
laws imposcd by the Amcrican occupation. Onc of the many ironics of
Japanese-Amcrican rclations is that hoth a ncw a ~ p r o a c htu labor relations and a new systeni of industrial linancc wcre imposed on Japan hy
Amesican occupation officials sympathetic tu Prcsident Franklin RooscveltZs" ~ e w~ e a l , "\vho liad been unablc to gain the political support
lor
measures in the Unitcd States. Two of R o ~ s c \ ~ ~ lmost
t's
vehement a n d ellective opponcnts in thc arca of labor law relorm ivere
Alfred Sloan and Hcnry Ford.
7. Toyota arid the othcr auto companics did employ considerable nilmhers
of temporary wurkcrs for many years as tliey struggled tu keep up with
growing dcmand and resisted granting life-time emplo),ee stalus lo
workers they wcre not sure they could retain. Howcver, this practice
camc to an end in the 1970s as thc Japancse Iirms gained conridcnce that
thcir groivth was not an accident and coyld he sustained.
8. The Introduction tu Michael Cusumano S, .lapanese Autor?zohile Iildiisif?
provides a succinct account of MITl's twcnt~r-ycarcffort tu rcorganizc
the industry and its failure to do so.
9. Intcrested readers are urged tu consult Ohno's work directly for tlic
details of his innovations: Taiichi Ohno, The Toyoia Prodriciion S?,.sient,
Tokyo: ~ i a m u ~ 1978
d , (in .lapanese). An excellent account in Englisli.
prepared with Ohno's help, is Yasuhiro Monden, 1-he Tuyovota Prodi~ctio~l
Wtenl, Atlanta: Institute of Industrial Engincers, 1983.
10. As we will s i c in Chapter 6 , a key additional problcm o€ this system 1in devising a workable bookkeeping system so that the real production
cost of in-liouse parts operations was known. Tu outside supplicrs, it
often appeared that arbitrary allocatiun of corporate overheads made
make/buy dccisions sham proceedings riggcd in lavor of the iri-honse
supplier.
11. Ohnc and Monden provide detailed explanations of this system in their
volumes un the Toyota Production Systcm.
12. we dcfinc a modcl as a vchicle with entirrly different cxtcrior sheet
metal lrom othcr producis in a producer's range.
286
END NOTES
14. Konosuke Odalra. Keinosukc Ono, Fumihiko Adachi, "Thc Automobiie
Industry in Japan: A Sludy of Ancillary Firm Uevelopmcnt," Oxfnrd:
Oxford University Press, and Tokyo: Kinokuniya, 1988,pp. 316-317.
15. Nishiguchi, "Siralcgic Dualism," pp. 203-206, providcs numerous examnles.
16. 11 is importan1 lo note that actual practicc differs considerably between
asscmblers. Tnyota designates two or more suppliers for mosl smaller
parts-for the front disk brake calipers for the base model Corolla, for
cxamplc. Nissan and Honda, by contrast, mainiain sevcral suppliers for
a given category of parts-for examplc front brake calipers in general.
Ho\vcver, Nissan and Honda du not dual- or triple-source a specific
part-thc caliper for thc front brakcs of a spccific model. lnstcad, they
assign onc specific part to each supplier and then compare their overall
performance. If a supplier iets down on a specific part, i i is relatively
casy lo transíer some 01' thc business lo anothcr suppliel- of thai type of
part. So in practice, thc Nissan and Honda sysiems are functional
equiiralents of thc Toyota systern.
17. Nishiguchi, "Compcting Systems of Automotive Components Supply."
18. Takahiro Fujimoto, "Organizations for Eífcctive Product Development,"
Table 7.1 Also sec, Figure 6.3 in this cliaptcr.
19. Nishiguchi, "Competing Systcms of Automotivc Componcnts Supply,"
p. 15.
20. FIis iindings are rcpurtcd in Richard Lamming, "Struclur-al Options for
thc European Aulomotivc Components Supplier Industry," IMVP Working Paper, May 1988; "The Causes and Effects of Structural Changes in
the European Aulomotii~eComl>onents Industry," IMVP Working Paper,
1989; and "The Inieriralioiial Autornotive Components Industry: The
Next Best Practice for Suppliers," IMVP Worlcing Paper, May 1989.
21. This survev uras conductcd by Susan Helper of the Bosiiin University
School of Management. Shc has reported hcr findings in "Supplier
Rclatinns a t a Crossroads: Results oSSurvey Rcsearch in the U.S. Automobile Industry." Boston University School of Managcment Working
Paper 89-26, 1989.
22. Richard Lamming. "Causes and Effects of Structural Change in the
European Automotivc Components Industry," pp. 22-23.
23. Toshihiro Nishiguchi, "Strategic Dualism," p. 197.
24. Susan Hclper, "Supplier Relations a t a Crossroads," p. 7.
25. Hclper, "Supplier Relations," p. 12.
26. Nishiguchi, "Stratcgic Dualism," pp. 116,203,204.
27. 1-Ielpcr,"Supplicr Relations," Figure 7.
28. Nishiguchi, "Stratcgic Dualism," p. 218.
29. 1-ielpcr, "Supplicr Rclations," Figure 7.
30. Hclper, "Supplier Relations," p. 7.
31. Nishiguchi, "Strategic Dualism", pp. 313-347, and Nishiguchi, "1s JIT
Rcally JIT?" IMVP Working Papcr, May 1989.
32. John Krafcik, "Lcarning from NUMMI," IMVP Working Paper, Scptember 1986; and Nishiguchi, "Strategic Dualism," p. 213.
33. Krafcik, "Lcarning from NUMMI," p. 32.
34. This section is based un Richard Lamming, "Causes and Effects of
Structural Changcs in ilie Eurupean Automoti\rc Components Industry."
35. Takahiro Fujimoto, "Organizations fur Effective Product Development,"
Table 7.1.
287
END NOTES
36, ~
~
~"Causes
~ and
i Effects."
~
g P. 39.
,
37. ~ ~ ~ ~ i ~ g , " Cand
a uEffects."
s e s P. 43.
CHAPTtR 1
1. Auionzolive Navs Mai*el Dula Buoii, various years.
2. John J. Ferron. "NADA'S Look Ahead: Projcct 2000," IMVP U'orking
Paper, May 1988.
3. A~~tornotive
,Vew.; Markei Daia Aook, various ycars.
4. ~ ~ l ~by ~the lauthors
~ t from
~ diluionrolivc Nei~9.s~rWuriiet~ a l a b o o k 1-89
,
editiun, p. 38.
5. John J. Ferron and Jonathan Brown, "The Future of Car Retailing," IMVP
Working Paper, May 1989, and Jonathan Broxijn, "What Will Happen tu
lhe Corner Garage?"
Brighion Polytechnic, Inaugural Lecturc, 26 Jliiic
1988.
6. SRI ~ ~ t ~ ~ ~ Tile
~ tFuiare
i ~ nfora Car
l , Dealeuships iil Eitrope: Esolution (Jr
Revolrtfion?, Croydon, U.K.: SR1 International. July 1986.
7. Ferron and Brown, "The Futurc of Car Retailing," p. !l.
8. Data suppIied by Prolessor Gai-el Khys of CardigBusln:ss Schooi.
9. This section draws on the work of Profissor Koichi Shimokawü rif Hosei
Univcrsity in Tokyo and on a case study o£ ihe Toyota distribution sysiem
prepared by Jan CIclling of Sxab.
10. The Auromobile Industq~:Jupuit arid X~yoiu,published by Tuyola hrlotor
Corporation, Tokyo.
11. Koichi Shiinnkawa, "The Study on Automottye Sales, Servicc and Distribution Systems and Its Furiher Revolutiori, IMVP Wurking Paper, May
1987.
~ " i ~ shimukawa,
hi
"The Study of Automotive Sales," p . 30; Auto~ilori%:e
Neius Murkei Daia Rook, and Japan Automobile Manulaclurers Association, iMoior I/ehiclc Srarisiics ofJapat1, Tokyo: JAMA, 1989.
Ferron aiid Brown, "Thc Fuiure o1 Car Rctaili~ig, pp. 4-5.
John J. Ferron and Jonathan Broxvn, "Thc Future of Car Rclailing," IMVP
Working Paper, 1989, p. 11, and private communicati<>nwith Kuichi
Shimokawa.
0thcr neccssary changes havc been the recen1 eliiriiiralion of special
excise taxcs on
~ ~ ilargc
t h engines (which penalized largcr imported
cars of the type with the most inhcrent salcs appeal in Japan) and thc
\villingncss of sc\rcral Japanese assemblers, under substantial international pressure, 10 x t a i l imports ihrough their own distributioii chann ~ l . ; --.
For <,xamnle. Honda has recently bcgun sclling Rover products in
Japan through'iis verno channel
CHAPTER 8
1. ~ l l a nNevins and Frank Ernest Hill, Ford: Decline aizd Rehirti7, Nciv York:
Scribner's, 1963.
are indebtcd to Maryann N. Keller of Furman Selz Magrr Dirtz an<l
2.
Rirncv .for
- an
~ cxnlanation of the Japancsc systcm of capital fr>rination in
the 1980s and f¿r the specific figures cited.
3. For an exeillent summary o[ the Ford Rilotor Co~ri~iany's
iorcign operations between 1905 and thc carly 1960s. see Mira Wilkens and Frank
290
END NOTES
3 . See thc "Mexican Auto Dccrce" promulgatcd by tlic government Dcccmher 19, 1989. in Diuvio Oficial.
4. For :i revicw 01' tlie Chinesc situation, sec Qiang Xue, "Thc Chinesc Motor
Iridusiry: Challenges b r the IYYOs," IMVP Working Papel-, May 1989.
5. For a revirw o l the Brazilian siiuation, see Jose Ferro, "Stratcgic Alternativcs lor tlic Brazilian Motor Vchicle lndustry in thc 1990s." IMVP
Worlcing Paper, May 1989.
6 . IMVP Rescarch Afiiliaie Jose Ferro visitcd ¡he Honda motorcycle plant
a t Manaus, lar up tlie A m ~ z o nnear tlle Peruvian border, and was amazed
a t tlie dcgree i r > which Honda had becn able io implernent lean production using rural migrarils ivitli absolutely no previous industrial cxperiencc. Tliis is surely ihc most diificult cnvironment in which lean produc{ionhas tlius far been tcsied and argucs strongly that the basic ideas are
truly universal.
7. This idea has bccn ioi-rnally acceptcd by Brazil and Argentina, but to
date tlie cliaos in botli economies Iias delayed scrious progress toward
iniplemeiilation. Fur a rcvicw 01' thc Argentine situalion see Javiar
Cardozo, "Tlic Argentine Automotii~eIndustry: Some Unavoidable Issues
Tor a Kc-entry Strategy," IMVP Working Papcr, May 1989.
8. This car finally rcached tlic Americari market in mid-1990.
INTERNATlONAL MOTOR VEWIGLE PROGRAM
SPQNSORING ORGANlZATlONS
1. For a rcview of the irnpact o l m a s s produciion on European tliinking, see
Thonias Mughcs, Amrvicuii Cenesis: A Cei?iruryo[Intsention urld %chnologicul Cizihirsiusnz. Ncw York: Penguin Books, 1989, particularly chapter 6,
"Taylorismus + Fordismus = Amerikanismus," and chapter 7, "The
Scciind Discoverv o l America."
AKZO n z
Australia-Department of Industry, Technology a n d Commerce
Automotive Industry Authority oI Australia
Canada-Departmcnt of Regional Industrial Expansiun
Chrysler Motors Corporation
Commission of the Europcan Communitics
Committee of Common Markei Automobile Constructors
~ a i m l e r - ~ e AG
nz
Du Pont d e Nemours & Co. Automoti\,e Prod~ictr
Fiat Auto SpA
Ford Motor Company
General Motors Corporation
Japan Automobile ManuIacturers Association
Japan Automotive Parts lndustry Associatiori
Mexican Association of ihe Automobili Industrv
Mexican Autoparts National Indusiry Association
Montedison Automotive Corporatc Group
Motor and Equipmcnt Manufacturers Association
Motorola, Inc.
Ontari-Ministry
of Industry, Trade a n d Technology
Peugeot, SA
Quebcc-Ministry of lndustry and Comrncrcc
Rcgie Nalionale des Usinm Rcnault
Rohert Bosch GmbH
Rovcr Group
Saah Cai- Divisio~i
Swcdish Nntional Board for Teclii1ical Dcvelopiiiciit
Taiwan-Minisirv of Ecoriomic Afiairs
'SRW Autonii:livc
United Kingdom---Dcpariment o l Tiadr and Inrlustry
United Kingdum-Econonric and Social Rescarch Crjuncil
United Stüles-Dcpartnicnt of Coirimcrce
Unitcd Statcs-Departrrierit oS'Transportaiion/NHl'SA
tinited Staics-Officc ol'T¿chnologyAssessmcnt
Volks\vagcn AC
Volvo Gar Coi-pora?i<iii
A P P E N D I X
I
I
Í lNTERfJATlOMW1 MOTOR VENICLE PROGRAM
/
RESEARCM AFFILIATE TEAM
Caren Addis, rMlT
Jonathan Brotvn, Brighion Busincss Scliool--Unitcd tiingdoiii
Javier Clirdozo, Univcrsit? ul Susscx--Uiiiicd Kingdon~
Matts Carlsson, Chalmers Univci-sit? o l Tcclini>liigy--S\iedcn
Al Chen, MIT
.Tocl Clark, MIT
K i m rlark. Har\,ard Busincss School
Michael Cusumano, MIT
Dermis DcsRosicrs, DcsRosici-s Automotivc Rcscat-ch-C;iiiada
Jose Robcrio Ferro, Uni~,ersidadFederal de Sao Cal-los-Bi-azil
lohn Ferron.
~, J D Po'iver- ünd Associatcs
Frank Ficld, MIT
Takahiro Fujimoto, H a r m r d Busincss Scliool
Lars.~rik Gadde, Chalmcrs Unii,crsity of T c c l ~ ~ ~ ~ ~ l c ~ g \ - S ~ \ ~ c d ~ ~ ~
Andrew ~ r a v c sUniircrsiiy
,
o l Sussei-Unitcd Kiiigdom
Susan Helper, Busto11 Uniitcrsit?
~
~
,
~ o h "~ e ~ w g oMIT
i,
Young-suk Hyun, Han Nam Uniicrsity-Korca
Masayoshi Ikrda, Cliuo Uiiivcrsit?-Japan
Daniel
.lunes. CardifTBusiiicss Schoul-Unitcd Kingdiiiii
TI-evor Jones, MIT Scnior Advis«r
flirister parlsson, Eurupearr Instituic for A d ~ a n c e dStuiiics
~
Iiarry Katz, Cornell Univcrsiiy
Hans IClein, MIT
Tliornas K o c h ~ nMlT
,
Jolin Kraicik, MIT
Donald Kress. MIT Senior
Advi.;<rr
~.~
- . ~
Richard Lamming, Brigliton Business Scli<ioi-United Kingdom
Richard Locl<c.bll?'
Jolin Paul ~ a c b u f . f i cMIT
,
Dcnnis Marler, MIT
Lars-Gunnar Mattsson, Stockholm School o l Economics-Sweden
Noah Mcltz, Uniuersity of Toronto-Canada
Gian Fcdcrico Micheletti, Politecnico di Torinct-Iial"
Roger Miller, Univcrsiiy of Qucbec-Canada
Trishihiro Nishiguchi, MIT
Kcniaro Nobeoka, MIT
,
lMVP Program and Forum Participanls"
-
Xiku Oshirna, Osaka City University-Jripari
David Ragone, MIT Scnior Adi,isor
David Robertson, MIT
Daniel Ri>i>s.MIT
Charles Sabcl, MIT
Slioichiro Sei, Karito Cakuin Uniuersity-Japan
Luba Shamrako\~a,MIT
Aritony Shcrilf, MIT
Haruo Shimada, Kciu University-Japan
Koicl~iShimokawa, Hosei University-Japan
Joscph Tidd, Univcrsiiy of Susscx-United Kingdom
Kontimi Tumisa!\'a, Long-Term Credit Bank-Japan
Kung Wang, National Central University-Taiwan
James IVomack, MIT
Victor Wong, MIT
Qiang Xue, MlT
UMBEK.I.O
A~,~El.~.~-Cllairman,
Fiat Auto SpA
Mexico
SHOICIIIA M E M I Y A - ~ ~ TGenera!,
~ C ~ O n'issan
~
bícxicaiia, S.A. de C .V.,
J ~ H BAll~A~-Dir~cto1N
General, Automorivc, Marinc aiid Rail Braiicli
(FAMR), Dci>artmcnt of Regional Ind~isirialExpansion, Giivcrnmcni os
canada
THEOBORE
BAI~OR-Presiden1o l Board of Directors. TEBO, S.A. de C.V.aiid
TEBO Group, Mexico
T. R. BEAMISH-Chairman, The Woodbridgc Group, Canada
FERNAND
B I I A I I N - D ~ ~ ~ General
C ~ U I ~ ior Interna) Market and Iniiustriai ALlairs, Commission of i11c Europea11Comnlunities
GIANCARLO
BEI~ETTA-Director,Automotive Corporate Group, Monicdisoii
STEPHENB ~ w ~ ~ - A s s i s t a nSecretary,
t
Departmcnt of Tmdc and I ~ i d ~ i s i r y ,
U.K.
MlcilAer. C n ~ ~ ~ < ; ~ i ~ ~ - R l l a nBusincss
a g e r , Stlateg)., Ford of Eirrripc, U.K.
CARi o s CALLEJA
P l ~ ~ ~ o - C h a i r m aofn tllc Board, Mac Elecironica, S.A. de
C.V., Mexico
MAIJRICIO
DE MARIAY C,ZMPOS-Subsecretaria de Fomento Industrial, Scci-etario d e Comercio y Fomento Industrial, Govcrnmcnt of blcxicu
FRANCOIS
CASTAIN-V~C~
Presideni, Vehiclc Enginccring, C h r ~ s l e rCurlioration
J A YCrrar-Execuiiipe Vicc Prcsidrnt, C. ltoh & Co. (America) Inc.
CHENZumo--Chairman, China National Automotive Industry Corporaiion,
Pe«plcs Republic of China
~
~
"Titles of individuals are shown as thcy \vere a t ihe time o l p;rriicipaiion in
IMVP.
296
APPENDIX C
J ~ ~ N E - SC
U ~I Io c ~ D i r e c t oIndustrial
r,
Policy Division, Minisiry oSTrade and
Industry, Repuhlic of Korea
MICHAEI.
COC~ILIN-UndcrSccretary, Departmcnt of Trade a n d Industry, U.K.
ROBERT
DALE-Managing Director, Automotive, Lucas Industries ple, U.K.
MICIIAEL
D~~ICGS-SpecialAssistant to the Prcsident for Policy De\~elopment,
Tile White House, Washington, D.C.
MANUELDE LA P~l~~~l.~.,l-Director,
C ~ m e r c i a lTransmisiones y Equipos
Mecariicos, S.A. de C.V., Mcxico
J O I I NEHY-Executixae Director, Corporatc Planning Oflice, Ford blotor Compan?
Do~ni.oEFIILIN-V~CC
Prcsidcnt, Internatiunal Uniori, UAW
Gus'rnvo E s i ~ r ~ o sCnliii~\~n~-Directi,r
n
General, Fabrica de Autotransportes
Mexicana, S.A. de C.V., Mcxico
CEStlR FLOKZS-Exec~itivePresiderit, Asociacion blexicana de Industria Automotriz, Mcxico
ANTONIO FKEIIO-Group Director, Finishes Division, Du h n t , S.A. de
Jos~
C.V., nilexico
PETEKIilI~it~-Mcmber of thc Board of Managemeni, Volkswagcl, AG
J~JLIIJ ALFKEDO
GENEI.G ~ ~ < c l ~ - D i r c cGcrreral
i~r
of tndustry, Secretaria de
Comercio y Fomento Industrial, Government OS Mexico
V~TKJKIO
G ~ 1 ~ ~ 1 . ~ ~ 1 - P r e s i and
d e n tChicSEsecu~iircOffjcer, Fiat Aiito S p ~
A i . . r x ~ ~ oG
e n~ n c c c r V i c ePrcsident and Cliief Executivc OSficer, ~
~
~
SpA
J o i i ~GILCI-IKIST-DirectorEjecutivo de Finanzas, Chrysler de Mexico, S.A.
KATHP-RINE
G r ~ ~ l l ~ ~ - D e p uProject
ty
Dii-cctor, Office of Tcchnology A
~
mcnt, U.S. Congress
GOROON
Griw-Depnty Minister, Minisrry oilndustry, Trade al1d Teehnolagy,
Govcrnmcnt of Ontario, Canada
J ~ H N
GRm-Executive Director, Corporate Strategy Staff, Fcwd Motor com.
pany
DONALD
GSCNWIND-ExecuiivcVice Prcsident, Product Development, chrys.
ler Corporalion
H E ~ GKl ~ s ? . ~ ~ s s o ~ - V iPresidcnt,
ce
Technical Relations, Saab-Seania AB,
Swcdcn
M I C ~ ~ A1-IA~~MES-V~CC
EL
Presideni, International Opcrations, Clirysler cor.
pnration
MICEIAEL
HAWLEY-Business Planning Associate, Ford ofEurope, U,K,
HIKOSHIH*~A~o-Prcsident of Honda of Canada M a n u f a c t ~ i r i nInc.
~,
KANH l ~ r \ s ~ l - P r ~ s i d e n New
t , United Motor ManuSacturing, Inc.
L o ~ ~ HUGE~ES-Vice
rs
President and Chief Financia1 Officcr, ~~~~~~l ~~t~~~
Europc AG
~V~AKSIN JOSEI'I+I-Pr-esidcnte del Consejo Ejecutivo, \.blksivagen d e Mexico,
S.A. de C.V., Mexico
KENICI~I
KATO-Director, Member of the Board, Toyota Motor- curporation
YosttltiA"~K a w n ~ ~ - M e m b c r oC thc Board of Directors, Group ~
i
~
Europe Sales, Nissan Motor Co., Ltd.
MAKYANN
KELLEK-Vice President, Furman Selí. ~ a g c ~r
i & ~i~~~~
~
t
~
JEAN-PIERKE
KEMPER-Director General, Automagncto, S.A. de C.V., R/lexieo
VON K I ~ I ! ~ E L L . - V Presideni,
~C~
TRW, Inc., Federal Republic
AI.EXANDER
Gcrmany
J o i i ~K~RSCI~IEN-Estcrnal
Relations Departrnent Director, Fiat Group Delegation with ihe European Community
APPENDIX C
LJI
s~~~~~
~ ~
~ ~ ~ ~ , Managing
\ ~ , , Director,
- ~ ~ Toyota
~ i ~RilOtor
~ r Corp«raiion
Deputy blinister, Ministrs of 1ndusti.v and Conimerce, Governrnent of Quehcc, Canada
~ - D ~Ministry
~ u ~ oflndustry,
~
Ti-adc alid Technolp A T ~ l CLt i , ~ ~ ~ ~ . L Ministcr,
"gy, Government of Onlario, Canada
R~~~~~~~ ~
~
~ and Chief
~
Exccutivc
~Ofliccr,h Rcgic ~
Na:ionaie
i des ~
Usincs Renault
J
~ ~ ~~
.~
~~
~~ Director-Dcsignate.
- f i ~ ~ ~E. 1.i d ~ Punt
i g dc Nemours &
Co., Inc.
cAnros
~8~Aun,lro-prcsidcnt,Corporaciori Industrias Saninis, S.A. de C.V.,
Mexico
J~~~ I
~ ~ , ~ ~ ~ A~ ~ General
~~ - D~ <if
j the
~ ~ ~Automotive
~ l ~ r Indusli-Y,Secret.alia
: dc ~~~~~~~i~
y fonrento Industrial, G o v ~ r n m e n os
i biexico
G~~~ ~,\O1.«M ~ ~ s , \ - s c ~ ~Vice
o ~ Presidcnt, Stratcgic Markctiiig, Fiat Auto
SPA
K~~ ~
~
~
~
~ scci-ctary,
~
~ .?i~toinotivc
~
- and
~ Chcmicals,
~
~
Depai-li
~
l
meni
~
~
dTcchnology
~
~ and
t Commcrcc,
~
~
, Governmcni of Australia
sADho
~ ~ ~ ~ ~ ~ Managing
~ ~ ~ Director,
- S eGeneral
~ ~ Manager,
i o r Internatiunal Opcrations, Akebono Brake Industry Co., Ltd.
jOHN
M ~ A ~ ~ Vice
~ Prcsidcnt,
< ~ ~ Automotii~c
~ - GProducts,
~ ~ E.~1. du~ i'ont
de Ncmours & Co., Inc.
~ E ~d , M~i ~ .~ s~ ~~ ~~ ~ ~~ Director,
~ ~ - Dirccspicer
LG
~~ S.A.
~ d e C.il.,
~ hlexico
~
~
~
nANs
~ ~ ~ ~ ~ , ~ Supervisory
- ~ h ~Board,
i ~ Rohert
~ a Boscli
n , GnrbH
pARVlz~ o I ; I . t . r A ~ l - ~ oVicc
~ o President
~a~
aiid Assistant General Manager,
~
~ Motorola,
~
.Irrc.
H~~~~~~~~
Moi;r~~-Director, Direciorate General 111 lriierna] E/farkel
and ~ ~ d ~ [ ~f a liCommission
~~~ ,i ~ ]of thc European Cominunities
H~~~~~~~~
~onruostti-President, Mazda R&D olNortlr America.
H~~~~~~~~~c
) CAST~~~o-ChieL
~
~ E x e c u~t i v ~Ofiiccr
~ and Director
~
Gcllci-al.
~
Diesel Nacional, S.A., blcxico
K ~ ~ ~ ~L, ~ .~ ~wr . s~- \ ~/Prcsident,
i r ~e ~
Cominission of thc Eui-oi~cüllC<immuniiics
~ l C l . l A K U~ ~ ~ ~ - p ~and
~ Managing
~ i d ~Dircctor,
n t General Mot(i1.s Mexico
ynsUsAo,, NoeuMmctChairman,
Japan Auto Parts Indusirics Association
R
~ ~ ~ ~ ~ ~ A ~ ~I~~ ~- d u~s t~rNacional
i~a - ~ ~ de ~Autopartcs,
j d e hrllxico
~ ~ ,
w~~~~~~
pnz CAsTirLo-Automotive Director, hrlinistry of Developmcnt, Go\'crnment of Venczucla
F
~ pER~
R I ~ . p ~ ~ ~ ~ rci ~ ~ - S~eGeneral,
cret~
a r yCoinmittee
~
of ColnInon
Markct Automobile Constructors
w
~ P E T E LK - S ~ ~ ~~O
~ i~r c c t o~t -Car
, Divisiun,
~
Daimlcr-Bcnz
~
~
AG
G ~ ~pEiIEuRA-~ice
~ , , ~ Presideni,
~
Champion Interamcricana. Lld. BuJias
Champion de Mexico, S.A. de C.V.
K~~~ H. P I I . ~ - ' ~ IG Metall, Federal Republic of Ger-illan?
w~~~~~~
~ ~ ~ ~ ~ ~ Motor
- pand~Equipmcnt
~ ~ i Manufiacturcrs
d ~ ~ t Associ,
~
l ation
~
~
,
Giteco~~
RAMPA-Chairman,
o
ANFIA, I t a b
H ~ ~ < ,R
, , ~ ~~
~
~ ~.c c r e~t a r y~,~ntcrnational
~
GMetalworkers
~
~
~fidcg-a~
~
tion, Switzerland
E~~~~~
~
~
~
~ and~Managing
~
~ Director,
~
- Chryslcr
p
~Mexico,
~
S.A.
~
i
d
R~~~~~
ti^^ Director, Corporate Strategy Sta[[, ford Motor
Company
L I ~President.
~\ZC
Condumcx Automotive,
ptoRo R~~~ M E ~ ~ O ~ A - E X ~ CVicc
Grupo Condumcx, Mcxico
M ~ ~ , , ~ ~
t
~
~
~
~
~
~
Gusrav RYDA~AN-Director,
?tilico and Industrial Deiiilopmerii, Sa;ib-\lil~net
AB, Finland
FeltNniino Snivcilrr UG~KTE-Secretariade Crinicrcii> y dc foiiicnti, Iiidustrial, Govcrnmcnt oí Mcxico
D O ~ ~ I N I SOIL\ vEE ~ - D i r ~ c L ~Plans
~ r , and Products, Peugcut S.A.
NICIIOLAS
S c l l ~ ~ L ~ - P r e s i d e aiid
n t Managing Dii-ccloi-, ki1-d Motor Ci>mpany
of Mexico
Louls S ~ ~ I ~ E I T Z E K - E x c c uVicc
t i x Prcsidcilt, Finaricc and Plaiining, Rcgic
Nationalc dcs Usincs Renaiilt
WII.I.IAMSCALES-Chairman, Autoii~otiveInditstl-y Autliority of Australia
GeRlInl¿o S c i i u ~ ~ ~ ~ t ! i ¿ - S e n i oVicc
r - President, Gcncriil Manager, Autoliiotivc and Indrlstl-ial Electronics Griiup, hlotoi-ola, inc.
H Y U NDON(;S I - I I N - E x c e ~ ~ tAdvisor,
iv~
Fiyundai Mocor Co.
WERNEU
SIEOEXT-ChieSEconomist. Volkswziagcri AC;
CI..E~I.IEN.I.E
~ l l i ~ O R o N l - ~ e ~ iVicc
i l ~ r Prcsidci-it, C o r p o ~ i t cDe\,cliipiiic~itand
Controiler, Fiat SpA
JDHNSnllT11, JR.-Presideni, General Motols Erii-ope .AG
J o t i ~S ~ e ~ l i e ~ s o ~ - R o vGroup
elHIDEOS u ~ i ~ . ~ ~ - A d v i sHiinda
ur,
Motor Compariy, Ltd.
TAKAO
Su%u~~-Cliicf,
Autolrl«tive Sectivn, Machiricry arid Ii-iíoi-nsutioii BLIreau, Ministry of lntcrnational Tradc and Industry, Japari
C ~ R I . - O ~ o i T ~ i t ~ n ~ u - A s s o c i ofS\\redisl~
ation
Autoiiiobile Manrii;rcti.irers and
Wh«lesalers
S ~ J ~ I C Touoon-President.
IIIR~
Japan Automobile Manufacturers Association
PETEL?
TUI~NUULL-hilanaging Director, Lcx Scrvicc pie, U.K.
DA NI EL.^ V~RDIAN~-D~~CCLO?,
Cvnlmission vi thc Europcan Comriiilriitics
roce^ V t ~ c E ~ ~ - M a l l a g i n1)irector.
g
Baiilcers Trust Coinpaiiy
R ~ O E KW,\TKJNs-~ep~lty
T
Assistant Sccretary, Automotive i\ii;iirs riisd Coiisumcr Guods, 1i.S. Department uf Cummercc
DANWEIIBIN-Executive Vice President, Vuliro Cal- Corpoi-ation
ROBER~S
WI~ITE-President. CAW-TCA Canada
MARINAWHI.~MAN-V~CC
Prcsident and Group Exccutivc, General Motors
Corporalion
JACKW I T ~ ~ ~ O \ V - E X Vict!
~ C UPresident,
~ ~ ~ ~ C Product Developmcnt, Chrysler
Corpuration
SHIGI?NOBU
Yi\~iihl~~<~-Chaim
Hino
l l , Muiors. Ltd.
YANGLi~ctiN-Orficial of Dcpartmcni o í Scicncc and Teclinol~ig~
Policy,
State Science alid Techniilogy Coinmission. Pcoplcs iicpublic <iíChina
YANGSHIH-CH~EN-DirectorGeneral, Industrial Dci~clopincniBureau, Ministry of Econumic Al'íairs, Republic o í c h i n a
T AYOKOYAMA-Director,
~
Dcputy Exccutii~eGeneral Manager, Office of the
Prcsidcnt, Mitsubishi Motors Corporation
CAI¿LOSZ A ~ ~ B K ~ N O - E X CDirector,
C U ~ ~ ~ Grupc~
~C
Industrial Ramircz, S.A.,
Mexico
ENKr~tleZAMURAN~-General
Dircctor, Mctalsa, Mexico
ZHUSUIYIJ-China Natiunal Automotive Industry Corporation, Pcoples Republic vi China
HaNs AHL~NDER-ProjcctManager, Purchasirig, búlvu Car Corporation
I'IEKOALESSIO-Fiat Auto SpA
DAVID
BECK-Managing Director, Lex Ketail Group Ltd, U.K.
MAUREEN
B ~ n R D - F ~ e e o ~ , % ~ - S e nPolicy
i o r Analvst, Department of Regional
Industrial Expansion, Government of Canada
AL B o s ~ ~ ~ - C h i cEngincer,
i
Enginecring Program Planning, Ciirysier Corporation
LAURESTA
BORSERO-Manager, Strategic Planning, Fiat Auto SpA
C t i e ~ Ziztrl-Seniur
,
Engineer, Chiria National Automotive Iridusiry Corporation, Peoplcs Repuhlic of China
HAKRYCoo~-Director, Automotive Research, Chryslcr Corporation
MJCIIAEL
DUI~E-SCI~~UT
Cunsultant, Arltomotive, Ministry of Industry, Trade
and Technology, Government o l Oniario, Canada
N ~ t i o i s nDIVLJAN-Director of Strategic Planning, Zastava, Yugoslavia
ELIE FARAH-Industrial Consuliant, Ministry of Industry a n d Commcrce,
Government oli Quebcc, Canada
M I C I ~ ~FEl hL- n e ~ s ~ ~ i ~ - A s s o c i Administrator
ate
for Research a n d Derrelopmeni, National Highway Trafftc Safety Administrativn. U.S.A.
R ~ B E RFTJ ~ z l i E ~ n ~ - V i Chairman,
ce
The Woodbridge Group, Canada
I\/IONT<;OMEIIY F R A Z I E R - D ~ ~ ~Sales
C ~ Uand
~ , Marketing, TRW Automotive
SWFT
~ O F
NR J E D M R N - R ~ S
Dircctor,
~ ~ ~ C ~lntcrnational Union, UAW
GERMA~NE
GIBARA-Aican Ltd., Canada
SAMG l ~ ~ l ~ - A s s i s t a ntot the President, Canadian Auto Workers Union
Sil1~JciiiG0.i.o-Group Manager, Toyota Motor Corporate Service of North
Ainerica
BASILH A R G K O V E - A S S ~tu
S ~the
~ ~ President.
~
Canadian Auto Vl'orkers Union
C ~ . n e s - G o ~HnE~~ . A ~ n e ~ - Q u a l i tManager,
y
Volvo Passenger Cars AB
J A NF~ELLJNG-Manager.Curporate Strategy, Saab Car Division, Saah-Scania
AB
MARKHOGAN-New Unitcd Motor Manufaciuring, Inc.
JGHNHo1.1.1~-Assistant Secretar? General, Committce of Common Markit
Automobile Constructors
D n r n o ~ iHoso~r-Senior
~
Executive Vice Presiderit, Carporale Liaison, S U baru of America, Inc.
JEANHouor-Deputy Director for Long-Range Planning, Pcugcot, S.A.
CANDACE
Howes-International Union, UAW
H n ~ s - V r c c oVON HULSEN-Secretary General and Chief Foreign Law Dcpartmenl, Volkswagen AG
v e President, Industria Nacional de
A ~ o ~ IsI ~
o A N EYZD u ~ ~ ~ ~ - E x e c u t iVice
Autopartes, Mexico
T s t i r o ~ uKAGAWA-AssociateDirector, Japan Automobilc Ma~iufacturersAssociation
STLIART
K ~ . ~ ~ ~ - D i r e c t oOffice
r,
of Automotive Industry Affairs, U.S. Department of Commcrce
REMI KELLY-Head, Automotive Division, Ministry of Industry and Commcrce, Go\,ernment of Qucbec
~
SUNGSHIN
KIM-Chiel Engineer, In-One Development Corporarion, Republic
of Korea
MINORU
I<IYOMASU-GeneralManager, Tokyo Research Dept., Toyota Motor
Corp.
R E ~ S I RKuRoMizu-Assistant
O
Corporate General Manager, Olfice of International Affairs, Mitsubishi Motors Corp.
GEORCE
LACY-President, Ontario Centre for Automotivc Parts Technology,
Canada
BOEI~JE
LENAS-Principal Administration Olficer, Planning Department,
Swedish National Board for Technical Development
MANUELA
LEROY-Assistant Secretary, Japan Automobile Manulacturers Association, Paris, France
En LEVS~~N-Scnior
International Economist. Motor Vehiclc Division, U.S.
Department o1 Commcrce
L1 SHouzHoNcDirector, Administrative Office, China National Automotive
Industry Corporation, Peoples Republic of China
HUAN-Vice Chairman, China National Automotive Industry CorpoLr Y ~ N
ration, Peoples Republic olChina
MARVINMILLER-Senior Research Scientist, Department of Nuclear Engineering and Center for International Studies, MIT
MUSTAFA
MOHATAREM-Directorof Trade Analysis, General Motors Corporalion
ALFRED MOUSTACCHI-Vice
President of Planning and Control of Investments,
Regie Nationale des Usines Renault
MARTINi i o ~ ~ l ~ ~ E L - V e h i cDivision,
ie~
Department of Trade and Industry,
U.K.
INDRA
NOOYI-Director, Corporate Strategy, Motorola, lnc.
JuorTii O'CONNELL-Business Policy Analyst, Automotive Industry Authority
o1 Australia
CHARLES
Ou-Chief of Transportation Section, First Division Industrial Development Bureau. Ministry o1 Economic Affairs, Republic o1 China
ROCERSP e ~ T ~ ~ s - E i e aoí
d Division, Interna1 Market and Industrial Aflairs,
Commission o1 thc European Communities
CAKLOTA
PEREZ-Planning Ollice, Ministry o1 Industry, Government o1 Venezuela
~ I A RPOM'ER-Vice
Y
President, Bankers Tmsl Co.
DORIAN
PRINCE-Interna1 Market and Industrial Aflairs, Commission o1 the
European Communities
GUALBERTO
RANIERI-Vice President, Corporate Communications, Fiat
USA Inc.
DAVID
REA-Director, Technology and Planning, E. 1. du Pont de Nemours &
Co.. Inc.
G ~ R DRICGS-Director,
~ N
Strategic Studies, Corporate Strategy Staff, Ford
Motor Comoanv
STEPHEN
ROGEICS-Director o1 Planning, Magna International, Inc., Canada
YOSHIAK~
SAEGUSA-VicePresident, Nissan Motor Co., Ltd. Head, Washington
Corporate Office
ROBEKTSAMARCC-Assistant Secretary, Automotive, Electrical Equipment,
and Consumer Products Branch, Dcpartment of Industry, Technology
and Commerce, Australia
SHINICHISET-Manager,
Operations Support Dept., Parts and Accessories
Division, Hinu Motors. Ltd.
A
,
SHENX~sr~-Senior Engineer, Information Division, China National u t o motive Industry Corporation, Pcoples Repuhlic of China
SHID l ~ ~ ~ u A ~ - D e p a r t m o1
e nIndustry
t
Technology, Statc Sciencc and Techiiology Commission, Pcoples Repuhlic o1 China
MOKIIIAIIU
S~rzuM~-GeneraiDirector. Japan Automobile Rilanufacturcrs Association, Paris, Francc
S i n w e ~Stio~uri~stii-Director,Automotivc Directorate, Department ol Rcgional Industrial Expansion, Government of Canada
Mnnti SNOWD~N-Booz
Allen und Hamilton, Paris, France
SreriitN SO~ERBERG-Partner,
Wellington Managemcnt Co.
NICOLESOLYOM-DEMESMAY-Deputy
Secretary General, Committee of Cummon Market Automobile Constructors
R I ~ H I \ RSTROI\.IBOTNE-U.S.
D
National Highway Tralfic Safety Administration
TAKEOT,IKAMI-DCDUIYGeneral Manager, Intcrnational Plannit~gOlficc,
Honda Motor Company, Ltd.
SEIJI T~N~\ti~-Directorand General Manager, Mazda Motor Corporatiun
Europe, R&D Representative Office, Federal Rcpublic olGermaoy
SrrrNrciir TANARA-Assistant to the Scnior Vice President, Corporate Public
Rclations, Amcrican Honda Motor Co., Inc.
BE~IC;T
TIDHULT-Principal Program Manager, Swcdish National Board lor
Technical ~evelop&ent
JAMESTI~ASK-Director, Competitivc Analysis, Economics Staff, General
Motors Corporation
FRED TocRE~-Generai Manager, Automotive and Industrial Electronics
Group, Motorola, Inc.
YOSHINOI~I
Usui-Assistant General Manager, Corporate Planning and Research Olfice, Toyota Motor Corporation
GERARDO
LOPEZVALAOEZ-Directorde la Industria Automotriz y Cuordinac i o n Secretaria de Comercio y Fomento Industrial, Government of
Mexico
STEPHCN
WALLMAN-Xhieí
Engineer, Powertrain, Voívo Car Corporation
AL WARNER-Director, Motor Vehicles Di~ision,Office of Automolil'e Industry Aflairs, U.S. Department of Commerce
FRANK
W H E L A N - C Engineer.
~~~~
Engineering, Resources Piallning and Control, Chrysler Corporation
JOHNWILLIAMSON-BusinessPlanning Associate. Corporate Stratcgy Officc,
Ford Motor Co.
DAVID
WORTS-General Manager, Japan Automobile Manufacturers Association (Canada)
MICHAEL
WYNNE-HUCHES-ExccutiveDirector, Automotive Industry Authority of Auitralia
KLNICHI
Y A ~ ~ s ~ l i t + G e n e rManager,
a~
Research Group, Ofhce of Corporate
Planning, Mazda Motor Corp
TOSHIAKI
Y o s ~ r ~ o - G e n e r aManager,
l
Sccrelariat, Hino Motors, Ltd
D
A P P E N D I X
(
IMYP PUBLICATII L
ISr
,14<jiiiliingProduciiorr Pi-<~ce.~ses:
Pasi Experience and Frlture Pluris
with Frank R. Ficld 111, Materials Systems Laboratory, MlT (Septembel1986)
Cost ibludeling ofi\lrer>ruiii~e
Aiitor,iohiie As.senih1y Techiiologies: A Coiiipuriitii'e Anulpis
wiih Frank R. Field 111, Yo~rngunLee, Jooncliui Park a n d Dehora11 L.
Thursion, Matcrials Scieiicc and Engirrecririg Dcpartmcnt, MIT (May
1987)
DcsRosiers, Dennis, DcsRosicrs Automotive Researcil, Inc., Toronto, Canada
,f Curiudiati Aitio~~roliie
Pclr1.s
The Si:c, Sirtictu,-e utiil Perf¿irmailce <itiie
fndusrrp: Iriel?rifiiizg Criticul Silcccss Fa'acturs (May 1987)
Ferro, Jose Roberto, Universidad Federal dc Sao Carlos. Sao Carlos, Brazil,
(Visiting Scholar, International Motor Vchiclc Program, MIT)
r$~i,iiun Ri?soriil-es Aiunngeirzent aiid Corpurute Czrlfcire E-ar~.sferiii !he
Briiziiiatz b1utor I/ehic/e Iiidiist- (Viewgra~hsonly) (Octohcr 1988)
Sti.ategic Alfernarii>es/¿ir rke Br.uziiiuii rl~lotorI/eizic/e Indcisrn? iii {he 1990s
(M- 19893
Ferron, John J., J. D. Powcr and Associates (formerly with thc National
Automohile Dealcrs' Assiiciation, Virginia)
Brown, Jonathan, Brightori Business School, UK
NAD/lS.sLook Aheud: Priijecr 2000 (May 1988)
Tke Francliised CurRerailiwg Iridiistry in ihe U.K. (Octobcr 1988)
Distrihritioiz Dynariiic.s (\fiewgraphs only) (Octoher 1988)
Tlre Fi~tz(reo{Cnr.Reiailirzg
with Johii J. Ferron, J. D. Power and Associates, USA (May 1989)
Tke F u f ~ i rofCar
e
Reiuilirig
with Jonathan Brown, Brighton Business School, UK (M-
Cardozo. Javier. Science Policy Research Unit, Univcrsity o[ Susscx, UK
The Argentirie Azitomotive Industry: Sonie Unai~oidableIssz~esfor a Re-enrry
Straregy (May 1989)
Carlsson. Matts, Chalmcrs University of Technology, Sweden
Cka1le11ge.sfor Organizaiion of Echnical Ftlnctions
ivith Christer Karlsson
Nrxt Pruclice ir! h4anagirig Producl Developrnent
with Christer Karlsson, European Institute lor Advanced Studies in
Management, Belgium (May 1988)
Next Pracfice i11 Produci Developnient: Integratio~rofEcknical Functions
witli Christer Karlsson
Clark, Joel P., Materials Systems Laboratory, MIT
"Thc listcd itcms werc prcparcd by IMVP rescarchcrs as part o i Program
rescarch. Most are available a s Working Papers. A Ccw items have also been
publisl~ediii jour-rials or- a s hook cliapter-s.
1989)
Field, Frank R., 111, Malcrials Systcms Lahoratory. MIT
ii/lodeling Productio,i Processes: Past Experieiice and F u f ~ l rPiuns
e
with Joel P. Clark, Materials Systems Laboratory, MIT (Scptember 1986)
Cosr Mudeli~igofAlieniarii,e Aiitornobile Asseinbly Techrzolugies: A Cornpurutive Aniilysis
itrith Joel P. Clark, Youngun Lee, Joonchul Park and Dehorah l.. Tliurston,
Materials Sciencc a n d Engineering Dcpartment, MIT ( & l q 1987)
Fujimoto. Takahiro, 1-Iarvard Business School
Tke 6ciopean iModel o{Prod~(c¿Dei.elopnie~rt:Ckiillen,se and OpporluriiQ'
with Kim B. Clark, Harvard Busiriess School (May 1988)
Consiste2lr Paztenis irr Autornotii~eProdzicr Struregy, Prodclci Deelopr??enr,
n>idMrinufaciuring I'erformance: Roud Map for the 1990s
with Aniony Sheriif, International Motor Vehici~.Program, MIT (May
19 89)
Gaddc, Lars-Erik. IMIT, Chalmcrs University o[ Tcchnology, Sweden
Ecirnical Dei~elopi~tei?t,
iblurkei Stnrciure Developi?leili and Distribution
Netivor!(s
with Hakan Hakansson. Uppsala, and Lars-Gunnar Mattsson, Stockholm
School r i f Economics, S~veden(Septcmher 1986)
Irzdusin D?'izar?iics iirid Di.sfiiburiorz
with Lars Gunnar Mattsson, Stockholm School oC Economics, Swcdcn
(May 1987)
Siubiiiiy iiiid Chaiige ii? Auioi~~oiiw
Distrih~irio>z
wiih [lakan Hakansson, Uppsala, Lars-Gunnar Mattsson and Mikael
Oberg, Stockholm School oT Economics, S\rreden (May 1988)
A Study of How Neiv Product Echnoiogies Are Adi~ppiedh?' Aliion~obile
Conipanies in ilze Uizited Staies, Et~ropeund.lapan (Sepiembcr 1986)
Hyun, Young-suk, Hannam Univcrsity, Korea (Visiting Scliolar, lnternaiional
Motor Vehiclc Program, MIT)
A Techizology Sirategy jor ilze Korran Motor. Iiidusiiy (May 1989)
Ikeda, Masayoshi, Chuo Univcrsity, Japan
Deaier Pei-speciiies oiz i2.1~inu/uctirrer.s'Total Perfbn7zaiice Straregirs
witli Lars-Guniiar Mattsson, Stockholin School of Economics, Sweden
(May 1989)
iln Internatiorial Conipañsoiz ojSt~hcoilii.acii~zg
Sy.sieit7.s i1i ihc Ariror?zorivc
Coinpnize~ftManufücturiizg Iwdrrsiq (May 1987)
with Lars-Gunnar Mattsson, Pcr Anderssoii, and Mikael Oherg, StockIiolm School of Economics, Sweden (May 1989)
U-Line Auto Paris Produciion
witli Shoichiro Sci, Kanto Gakuin University, Japan, and Tosllihiro
Nishisguchi, International Motor Vehicic Program, MIT (Octobcr 1988)
Graves, Andreiv, Science Policy Research Unit, Univcrsity of Sussex, UK
The .Tupanese Auio Conrponenf Maizr<facturers' Sysierii jor rhe Dil~isioiziif
P r o d ~ ~ c t i(Mdy
o i ~ 1987)
Eclinolí~gyCllui1eiige.s Fucirig ¡he M ~ ~ i oIiidusiiy:
iRight und Wrung Sirategies (Septembcr 1986)
Tlie k n s f e r of. Flexible Production Sy.steii~stu Japanese A~itoParrsinakrr
Pan.splant,s in !re U.S.
with Shoichiro Sei, Kanto Gakuin University, Japan (May 1989)
Coniparison ojInterriationul Reseurch aizd Dei,eloprneizr in the Autornohile
Indiisiry
with Daniel T. .Tones, SPRU, University of Sussex, UK (Septcmber 1986)
Jones, Danicl T., European Research Director, Intcrnational blotor Vchiclc
Program, Cardiff Business School, Univcrsitv o l Wales, UK (fi~rmcrlyof
SPRU, University of Susscx, UK)
Co~?zpurativeTuends in Auiui7ioiive Research and Dewlvpnzent (May 1987)
(rc\,ised Septcmber 1987)
The Dyilumics of the Worid ,h4oror Vehicle Iridn.si~:Irsuer jor Aiiuksis
(Septcmber 1986)
Ei~ropeuir Deugn uiid Engrneenng Capabrlitre:, A Coizrrrruiiig Strengih
(Mdy 1988)
Coinpurison of1,rternarioirul Re,seurch aiid Developirieizr iiz tife Aiitoirlobile
Industiy
with Andrew Graves, Scicnce Policy Rcscarch Unit, University of Sussex,
UK (Septcmber 1986)
B r o ~ ~ ~ ~ f iTrunsplaizis
eld~,
und Neiv Enrranis: Tke Oiw'c<ipucil? l1>vh1i.~7i
(May 1987)
Design Hoiises aizd the Iiziroduciion ojNei~iEchi~ology:A Case Strrdy (May
1988) (revised Octohcr 1988)
Echnol«gy Pends in the World Arrtonzohile Iizdustq' (Viewgraphs only)
(October 1988)
Depurrrire iiz Auroniobile R&D? (May 1989)
Proinethens: A A1e%<i
ticlper, Susan, Department of Operations Managenleni, Boston University
Changing Stipplier Relutioii.ships in ihe lis. Arrto Ii~drrstry:A Fraineivork
for Anulysis und Proposal jor Survey Researcii (Octohcr 1988)
Chunging Mippiier Relutionships iii rhr United Staies: Resnlrs o j Slrrvey
Research (May 1989)
Merrigel, Gary, Dcpartment ot Political Science, MIT
Collahorative Mrrnufkciuring: New Sripplier Relatiolrs in the Autoi?iobile
Indusiry aiid the Rti<lefiizitionofrhe Industriul Corpvrafion
with Charles F. Sabcl, Departnient o€ Political Scicnce, MIT, and Horst
Kern, University of Gottingcn, West Gcrmany (May 1989)
Strrrciural Adjusli?zeni in tlie Acrioinohile Indusiry
STIReview, OECD, No. 3, April 1988.
The Cornpetitiix Posirion o/. ihe Eirropeaii ,Motor Indz~srry:ilie Ruce jor
Added Valrre (May 1988)
Measuring Zchnological Adwritage ir1 the Moior ~ l z i c i eI i z d u s l ~(May
1988)
Key Findiirgs of the MIT Interi~utionalMotor llehicle I'rogrui?? 1987188
A Suinrnary of the Reseurch Poiicy fircim, Spriizg 1Y88
with Richard Lamming, Brighton Busincss School, UK
The Neiv Entraizis: Searcliing for a Role ir1 tize Worid
with James P. Womack, Rescarch Director, lntcrnatioilal Motor Vcliiclc
Program, MIT (Ivlay 1989)
A Second Look ui rhe Eiiropenn Motor Industq' (M-
Hcywood, John B.. arid Viclor W. Wong, Sloan Automotive Laboratory, MIT
i h e Errmpeair Motor Ii~dusrryaizd.lupuri iiz tile 1990s
1989)
306
APPENDIX D
The JAMA Forum, Spccial Issue on the Automotive Industry and EC
Single Market (September 1989)
Corporaie Strategy and iech~iolog~'
iiz the Autonr<ii,ileIndirstn,
in Mark Dodgson, cd., Eclznology Stralegy ami ¡he Finiz: Managenleni and
PzrblicPolicy, London: Longman, 1989.
The Conipetitiiie Oiftiookfor- tlre Europeurr Auio ?zrdi~stInternaiiorial Jouniul ofEhicle Desigi~,May 1990.
Meusuring lip to ¡he Jupanese: Lessons froiu ihc Abror Irid~istiy
Universiiy of Wales Business and Ecoiioniics Reljiew, No. 5, 1950
Karlsson. Christer, European Institute b r Advanced Siudies in Management,
Brussels, Belgium
Challeizges for O~y.a~iizution
ofEcIznica1 1:iiiicriun.s
with Matts Carlsson, IMIT, Chalmers University of Technology, Sweden
(May 1987)
Nexf I'rflcrice in Maririgi?igPIndrict Dei~rlopme~zt
with Matts Carlsson, IMIT, Chalmcrs University of Technology, Sweden
(May 1988)
Next Pruciicc in Produci Deveioprnenr: Iiilegrarioi~of Eclrnicul F~uictiorzs
with Matts Carlsson, IMIT, Chalmers University of Tcchnology, Sweden
(May 1989)
Katz, Harry C., New York State School of Industrial arid Labor Relations,
Cornell University
307
APPENDlX D
Krafcik, John F., International Motor Vehicle Program, MIT
Leanzing from NUMMI (September 1986)
Comparative Munr~jaciuilngPraciice: Iinhalunces uiid Inip1icuiiori.s
with James P. Womack, Research Director, International Motor Vehiclc
Program, MIT (May 1987)
Trei~dsiri /izten?nii»nal Auioniotive Assernhly Prucrice (Septcmber 1987)
Comparative Aiialysis u j Petfomzar?ce I ~ l d i c a i o r(ri
. ~ World ilrrio A.ssei,~h!s
Planrs (Master of Science thesis, Sloan Scliool of Managemeni, AIIT,
January 1988)
European Mannfaciuriizg Praciice iii u U'u1.1d 1'er.speciiiie (iWay 1988)
Cornplulity arid Fieiiihiliry iiz híolou Vehicle As.sen~biy:A Ltbr!divide Per.?pective (May 1988)
A Sutlirizary of Fi~rdingsaunil Fuiure Reseiirch in hJa~iufuct~<i-ing
Pructice
(October 1988)
A Meihodology fijr Assemhiy Piunt Perfimruirce Deteiiiiiiiuiiun (October
1988)
Tlze Problern of Flexibilify for the S~rpplierIrrdusi~y:Aii As.serirb!i. Plurlt
Perspectiiie (October 1988)
E.xolaii~ingH i ~ hPerf¿orn?anceManufacturirig: The I ~ z i e r n u t i o ~Aut<>i?iu~al
tive Assekb/y Plan¡ Study
with John Paul MacDuffie, International Motor Vehicle Program, MIT
(May 1989)
The Ii~dustrialRelatiorls Clrullenges Fucing ilie Wc~ridAiuo Indust-i (May
1987)
Assembiy Plant Perfbnlzai~ceand Chunging Markei Stnicii<i.e iir ihe Li*ri*ql
Car Segrnent (May 1989)
Effects oflndusfriai Relaii»izs on Prudt<ciivip:Ei~iderzcef h i n a12 Aniei-ican
Auto C o m p a q
with Thomas A . Kochan, Sloan School of Managcment, MIT (May 1988)
A Coinparative Analysis of Assetnhiy Plarzi Auioiizaiion IMay 1989)
Changiiig Work Practices uild Prodt~ciiviiyiiz the Auio Indi<sirv: A U.S..
Canada Cornpariso~z
with Noah M. Meltz. University of Toronto, Canada ( f r o n ~Iizdusiriai
Relations Issues for ¡he 19909, Prvceedings cif the 2óth Cvufcrcnce of the
Canadian Industrial Rclations Association, Juiic 4-6, 1989, Laval University, Quebec)
Klcin, Hans, International Motor Vehicle Progiam, iVllT
Tobvards a U.S. Nuiionul Progrriin in Inielligeizi Ehic/elIi'ighivui> Sysienis
(May 1989)
A Fir.st Look ar
1989)
Performance Levels ai Neiv Enirani A.sser~iblyPiunts (May
The Earn Concept: Models joi- Change
with John Paul MacDuffie, Intcrnational Motor Vchiclc Program, MIT
(Tlie JAMA Foru~n,Vol. 7, No. 3, February 1989)
A Neto Diei for U.S. Ma~irrfucturing
iechiroiog)' Rei~ieiv,MIT, January 28, 1989
Tr~nmphof the Lean Produciioi~Syster?~
Sloan A4anagemeut Rev~eii,,MIT, Vol 30, No 1, Fa11 1988
The Effect o{ Desigiz Murrnjuciuruhilify o17 Producii;ziiy und Qtiaiii~':A??
Update on (he IMVPAssenibly Plailt Siuds (Januar), 1990)
Kochan, Tliomas A., SIoan School of Managcnient, MIT
Effects of Industrial Relatio~iso17 Producrii>iij,:Ei,ideiice froin a,? Aiiiericun
Ar~roCo~naanv
with Harry C. Katz, Ncw York Stalc School of Iiidustrial and Labor
Rclations, Cornell University (May 1988)
Kress, Donald L., Scnior Advisor, Intcrnationai Motor Vehicle Program, MIT
Iinplications
1988)
of Ecropeun Uriificaiion for the hfoior Vehiele Ind~isfiy(May
K v 1s.sltes in Moior Eliicie Disiribi<iioii(May 1988)
308
APPENDIX D
L a m m i n g , Richard C., Brighton Business school, U K
The Iiztematio~zal Autoniotive Components Industry: C~ofomer-Supplier
Relationships: Past, Present aiid Future (Mav 1987)
Changing Relationships in rhe North Americaiz Autoinofiw Components
Indrutry: Norfh AmericanlEuropean Perspecfixs (September 1987)
Strilcturul Optiorzs for the European Autoinotive Coi~iponenrsSupplier
Industry (May 1988)
The Intenzaiional Autornotiine Compoizenfs Supply I~zdustry:The E~nerging
Best Pi-acfice (Vicwgraphs only) (October 1988)
The Post Japaizese Model for Inteniatioiiul Autornotiie Compoizents Supplj
(October 1988)
Tier Structzlres iti /he Norrh A~?ierica,iAutomotii,e Co~nponenisIndu.stv
(October 1988)
Key Fiizdings of the MIT intenzational A4otor Ehicle Progra~n1987188
w i t h Daniel T. Jones, Cardiff Business School, University o[ Wales, U K
(October 1988)
The Iizternutionai Auton~ofiveComponenfs Supply Iwdustry; The Enzergir~g
Besr Practice (October 1988)
The Inremalional A u t o ~ n o t i wComporzentr Indllsfry: The Next ' B e s t Praclice" for Supplien (May 1989)
ve
Supp1ier.s ofNew
Research and Dexlopntent in the A ~ i f o ~ n o t i Conzponents
Entran1 Countries: The Prospects for Meiico (May 1989)
The Causes and Effects o f Structural Change i n rhe European Automotive
Components Industry (1989)
MacDuffie,John Paul, Sloan School o f Manaeement. MIT
Industrial Relatiolzs and "Hunzaniuare": Japanese 1nvestinent.s in Automohile Manufacturing iiz the United Srates
w i t h Haruo Shimada, Department o f Economics, Keio University. Japan
(September 1986) (Revised for May 1987 meeting)
The I~zteractiono f Productioiz Methods, H t f m a n Resotlrces, und Techizology
in Maizufacturing Practice (October 1988)
Expluining High Performance Munufactilring: Tlie Inten~aiionulAutomotive Assemhly Plaizt Stndy
w i t h John E Krafcik, Inteniational Motor Vehicle Program, MIT (May
1989)
IUorldwide Peizds irz Producfion System Managenient: Work Sysfern.s, Factory Practice, and Huniun Resource Managernerzt (May 1989)
The Team Concept: Models for Clzaiige
w i t h John F. Krafcik, TheJAMA Forunz, Vol. 7 , No. 3, Fcbruary 1989
Marler, Dennis L., Intcrnational Motor Vehicle Program, MIT
The Post-Japanese Model ofAuromotiiie Cornponent Supply: Selecter¡ North
Amencan Case Studies (May 1989)
APPENDIX D
309
Mattsson, Lars-Gunnar, Stockholm School o f Economics, Sweden
Technical Develo~mewt,Market Stmcture Development and Distrib~ttion
Networks
w i t h Lars-Erik Gadde, IMIT, Chalmers University o f Technology, and
Hakan Hakansson, Uppsala,Sweden (September 1986)
I>zdustryDyizamics and Distriblction
w i t h Lars-Erik Gadde, IMIT, Chalmers University o f Technology, Swcdcn
(May 1987)
Stabilily and Change in Autonzotive Distributio~z
w i t h Lars-Erik Gadde, IMIT, Chalmers University o f Tcchnology, Hakan
Hakansson, Uppsala, and Mikeal Oberg, Srockholm School o f Economics, Sweden (May 1988)
Dealer Perspectives o n Maizufacturers' Total Perfomzunce Strulegies
w i t h Lars-Erik Gadde, IMIT, Chalmers University o f Tcchnologv, Sweden
(May 1989)
Reorgaizizing Disrrihuiiorz for Total Perfomzance-The Manufactnring Vieivpoini
w i t h Per Andersson and Mikael Oberg, Stockholm School n i Economics,
and Lars-Erik Gadde, IMIT, Chalmers University o f Technology, Swedcn
(May 1989)
Meltz, Noah M., University o f Toronto, Canada
Chaneinz Work Practica and Productivity in the Auto Iiiilustq~:A U.S:
~ a n a ; i aEoniparison
w i t h Harry C . K a t z , Cornell University ( f r o m Industrial Relatioizs Issnes
for the 1990's, Proceedings o f t h e 26th Conference o f tlie Cariadian
Industrial Relations Association, June 4-6, 1989, Laval Universily.
Canada)
Micheletti, Gian Federico, Politecnico di Tnrino, Italy
Future Trends in Process Technology: Quality aiid Reliability Iinproveine~?t
with Autonzatic Productioiz (May 1988)
Miller, Roger, University o f Quebec, Montreal, Canada
N a v Locational Factors i n tke Automohile Industry (October 1988)
The New Locatio~zalDynamics in the Automobile Irid~lstry:Assenzbb Fucil[ties,Parts Plants and R&D Centers (May 1989)
Competitiw Dynanzics and RdiD: The Locatiolzal I m p a c u (1990)
Nishiguchi, Toshihiro, International Motor Vehicle Program, MIT
Compering Systems nf Automotive Conlponerits Supply: An Exaininatioi? o f
tlze .lapanese "Clustered Control" Model and the "Alps" Structnre (May
1987)
Netv ii;olids i n American Auto Coniponeizfs Supply: 1s Good hfu~zage~nent
Alivays Cnlturally Bound? ( S e p t e m b e r 1987)
310
APPENDIX D
Refoming Auromofive Pnrchasing Organizarion ir1 Norfh Anzerica: Lersons
fui. Europe? (May 1988)
U-Line Aciro Parfs Production
with Masayoshi Ikcda, Chuo Univcrsity, and Shoichiro Sei, Kalito Gakuin University, Japan (Octobcr 1988)
1sJ I T Reul1.v JIT? (Rilay 1989)
Tlze Jupunizafion ofthr U.S. Auto
Good Ma~zugemeni1s Guod ~Manugenie~tf:
I~idust~
The JAMA Forum, Vol. 7. No. 4, April 1989
O'Donnell, John P., International Motor Vehicle Prograrn, MIT (also of the
Transportation Systems Ccntcr, U.S. Dept. of Transportation)
Cornpeririie Pmducr Programs and Anticipated Doinesfic P~odncrioriaiid
Auto-Relared Einployrnenf for 1990 (Scptemher 1986)
Brownfelds, Ti.an.sp1anr.sa n d Ne~vE~irralzrs:Tlie Oi~ercapaciryProbleii~(Tlie
U.S. Per.specriie)
with Laurie Hussey, Transportation Systems Center, U.S. Department of
Transportation (M- 1987)
Addendum to Broivnfelds, Truizsp1arir.s oizd Netv Enfi.anls: Tlie Overcapaciry
Problem (The U.S. Percpecrive)
The Norrh A>nericaiiLighf Puck Murkei
with Laurie Hussey, Transportation Systems Center, U.S. Departmcnt o[
Transportation (Scptcmher 1987)
~
A Second Look ur Develupnlerirs in rhe U.S. Iiidu.srry ( b 1 1989)
Oshima, Taku, Osaka C ~ t yUniversity, Japan, (Visiting Scliolar, IMVP, MIT)
S t n ~ c i i ~ rCoiiiparison
al
of fhe Jupuiiese and Cliinese Auroinohiie Irid~islries
(Septemher 1987)
Zéchriology Trunsfer ofJupane.se Acironiakers in rhe Uniied Staics: ,Muzda
Motor Corporation Case Study (May 1989)
Robertson, David, Intcrnational Motor Vchicle Program, MIT
CAD Systeins in the Desigiz Eiigineei-ing Procesi (May 1989)
Sahel, Charles F., Dcpartment of Political Scicnce, MIT
Collahoraiive hJanufacfuriiig: Neiz~Stlpplier Relations in rhe Auroniohile
Indci.st~yaiid rhe Rede/iiiiri«ii offke Iiidusrrial Corporuiion
with Gary Herrigel, Departmcnt of Political Sciencc, MIT and l o r s t
Kern, University of Güttingen, West Gcrmany
Sei, Shoichiro, Icanto Gakuin University, Japan
The Elecrr»nic JIi' Sysfeni and P~.oductionZ;!chizoiogv (May 1987)
U-Line Auto Parls Producfioii
with Masayushi Ikeda, Chuo University, Japan, and Tosiiihiro Nisliiguchi,
International Motor Vehiclc Program, MIT (Octobcr 1988)
3111
APPENDIX D
~+<~diiciioiz
Sysreins to ./apaliu>leseAzrto I+irisrtiaker
rlie ~ n i l i ~ ~ iI;icxiijie
.
Triirrspl<i>~l.s
irr tlie L1.S.
\vitli Riiasapshi lb<:&, Cliii<iUiii\,ersiiy, Japan (May 1989)
Sliarni-akora, 1-uba, Ciinsuliant, Inicrnational Motor Vcliicle Program, MIT
The Sui;iei ilrlioii~<irii~c
Iixliisiv iiiid !Ciiiier in Liglzi
Re/oi->l?.s(May 1989)
afilie Neiv Ecu~lotizic
Shcrifl., Anir~ny,Irriel-natiaiiai Motor- Vcliiclc Prograrn, MIT
?.he Col?liietiii~:eProdiict Pc>sirioiiof.iiitioirohile ;i/lui~i~ii<~ciu~~~v.s:
Perjoi-111<iticeniid Si~aregies(Ma? 1988)
~ ; ~ , ~ r e Iini~/icaiioiis
gic
io tlle Auto itid~isri:+
j.jfc cycie.; ritld
1Octobcr 1988)
T/IL!F!:<igiiie~~r(itio,i
of tlie I,L4ii./d !!~liitur l~klticic1\1ai-/<eia ~ Ir.?
d I'ore~zfial
Itiipaci o>iilic S~ipplieuI i i d ~ i s t v(Octobcr 1988)
Coii,si.sreiir I'uriei-lis ivi Aur»i?iotit:e P i i i d ~ ~Srruregy,
cf
Prodtlcr Developmei?i,
riird ,\,Iaii~ifiiciur-ir!g?>eiT¿~i-tniillce:
Roiid Mup /¿ir i/!c 1991')s
\vilh Takakiiro Fujirnoto, ffai-vard Busincss Scliool (May 1989)
Sliimaiia. liaruo, Dcparlncnt of Ecoiiomics, Keio Unixzcrsity,Jal?an
(iil Japanesc), Tokyii: Dialnond. 1989
Tlie Ecorz~rnic.s~fNiui?iitii!~ai~e
~~
~ , ~ d ~ ~, ~ ~j ~ il ~ )~"ffiilniir~i+~nre":
~
i
i J i i~p a r ~~c s1tim.s'ri~zeizrs
~
~in Aiifo~nohile i\iaiz~~fucrui-iiig
irz ihe IJriireiI Sfares
i k i i i l i 1rihr-i Paul MacDuCiic, Sloan School of I\ilanagenlcnt, MIT (Scptem;lrett.
ti<!,st,
~
~ atid H~iriliin
~ ~esoui.i;e
~
Srruiegies:
~
ii Chullciige
~
{br
n rlze Jupa~
..liiioniobi/e Irid~i.si~y
íMay 1989)
Shiinokawa, Koiclii, Hosci University, Japari
Tlic Srir(iy o/Auroinoii~-cSule.s, Disri-ihiirioti uild Seiivice *sieiiis aud 12s
Fiiuilirr Rei~olrriiorz(May 1987)
Viializ.iiig Ainori~ohilcSiiles @.siein ?Oi<,urd,sDe-iblairircd Age: Chaiigii?g
Japiiriesc Aiiioinohile iI.la~.ketuizd Receiif Experiences of.Iapnnese ~ ~ l f 0 1 7 ~ l i hile Miiiziifacriirev.s (M- 1988)
Uc~i~eloixii~~r
ofrhr A.siit~iA1lL.S A~tfotnohiieIiiíl~isriyaud Fi,tfnre Pio.cpeci.5
o f f l ~ (;lohui
e
Dii,i.sioii ofl..rihor-Jripail, ROK, Chiiia-Tuiit'iiil atiii Tiiiiilund
(Rikry 1Y 89)
Tlic Firiure o/. A ~ ~ r , ~ ~ ~ Di~frihitii~iz:
iol~ile
Rei~oliiiioii aizrl Retiiett, o[ tile
I~
Cun.scri?ieuNeeils
Alif<i,7i<lhil~
Siile.s (lnd Di.sri.ihriiio11S ~ S I PU,i<ier-Charigiiig
<iiiiliLla!:kcr Slilictrire (May 1989)
T i & , Josepll,
Sciellcc I'olicy Rescarch Unit, Uriiversit), uf Susscx, U.K
A'mi Siep.7 iii /lsseinbIy Airioi~?<itio>i
(May 1989)
Toniisa\va, Kr>ri<iii?i,
Isiiig-Tcrnr Credit Bank, Lid., Tokyo, Japari
i
312
APPENDIX D
Dei~elopinentof and Ftlture Outlook for nii liztematiolza! Dii~iiiono/.Labor
ir? the Ai~tuniobileind~i,strie.sofAsiar? NlCs (May 1987)
Wang, Kung, Dept. o f Business Administrarion, National Central University,
Taiwaii
Devei»pi?zeilrStruregies fbr tize A~ifornobileand Paris Iizdzisiq~o f the Republic ofChbiu (May 1989)
Womack, Janies P., Research Director, Internatirinai Motor Vehicle Program,
MIT
Cross-Nutioizal Col1uhnruriorz.s irt rhe ~lSoi»r-lndusliy: Theii- Ca~isesund
Con.sequeizces (Scpteinber 1986)
ilSoviizg Toii>u~-d
ih41rliiivideRest Pructice: Critica1Citoices for C~~untries
rind
Conzpanies (May 1987)
Comparatit>cManufucrunr~gPractice: Imhalarices aud Iniplicriiions
witli John F. Krafcik, Intcrnational Motor Vehicle Pn~gram,MIT (May
1987)
The Fuiure Sltupe ojthe World Auto Irzdrrstly: Retiiiuking Indusrn Siniciure
(Septeniber 1987)
Collahorution as u Struteg), for rlchievir~gBrsi Prriciice (September 1987)
The Seurch for Best Pructicd (September 1987)
Mulrina~ioizulJoir?iI/enrt~resiil the Motor I/ehicle Sectov (1987)
Prospects /o?, the US.-Maicari Relario1ishi.p iir the Moror VEhicle Sector
(1987)
i'hr Deiu?lopinerzfo f rhe Chirresr Motor &hiele Indusrn: Straiegic Alternalives und the Roie ofForeign Firr?t.s (December 1987)
The Etiropean iC111to1.
Indnstiv in o MJorld Co~ztmt:Soine Str-aiegic Dileinrnas
(May 1988)
A Revieiv oflhe 11blVPReseurch Prograni (Vicwgraphs only) (Octobcr 1988)
Tize Stute ojthe Arilo Iildusin': hloving ro fhe "POSI-Japurzese"Era (Ociobcr
1988)
Struregies fijr a POSI-NutionalMotor Ii?diislt?~(May 1989)
Tlie Neiv Enrrarirs: Seurching f ¿ ~urRole iw rhe Wb,ld
with Daniel T. Jones, Europcan Rescarch Director, CardiCf Business
Scliool, University of Wales, UK (May 1989)
Tlze Mrrican iM»tor I n d ~ i s t yStra1egie.s for thc 1990's (May 1989)
A Posi-Nuiionul Auto Industi?' by the Euu 2000
The JAMA Ii~runz,Vol. 8, No. 1, Septcmher 1989
Wong, Victor, Sloan Autoinotive Laboratory, MIT
313
APPENDlX D
with jOiln B. ~
(Scptember 1986)
~ Dcpartmeril
~
oi Ivlcclianical
~
~ Engineering.
~
Mil-d
x u c . ~ i a n gfnternational
,
Motor Vehicle Prograin. MiT
o,,iiine o f ~ e s e u r c h lhc Chiuese hfiitor \@iiicie[iidi~sirr(Ociohcr l'J8K)
rhe clzitIese,%4oioy~ , i d ~ l , s tClinl1oige.s
v:
/¿>rrhe 1 Y9U'.s (Mri? 1989)
,
AC Spark Plug, 138
Aggrcssive selling, 67, 186
Agnclli, Giuvanni, 44,231,234
Agriclli farnily, 197
Aisin Sciki. 195
AlFa Romeo, 121, 193
Arnerican Molors, 214
Ar-iiold, Hurace, 28, 32
Associiiiioii of Soutlr Easi Asian
Nations (ASEAN) countries, 271
Asion Miirtin, 25, 51. 65, 229
Ailiinia Ford Assernbly Plant, 96
Austin, llcrberi. 44,231,233, 234
Austin Motor Cornpany, 231
Australia. auiomobilc indusirv in
270-72
Autornohile industi-v. See üiro Eumpeair automobilc iiidustry; Japancse autornohile industry; L1.S.
üui<irn<ibiicindustrv
in Auslraliri, 270-72
in Brazil, 269-70
a s cyclical, 42-43
ir] ileveloping countrics, 263-70
in East Asia, 267-69
Europeiin innorations in, 46-47
in Mexico. 87,240-41.264-67
o\rerclipacity crisis in, 12
productiun by regioii, 43.44
union rno\,ement in. 42
315
Automotive ;issciribly p!ai?ts, 73--103
ciassic leal! prijduciioi~,79-80
classic m a s pi-odiiction, 77-78
cornpzirison id inass ;ind icain produciiori, 75-77. 80-82
ciaft pri~duiiion,88--~91
in drvcloping cuun1:-ie;, 87
dilfusiiig iean prudrictiun i i i , 82-84
impruving, 93-98
IMVP sui-vey u!, 84-88.Y 1.~93
lcan orgcnizatiuii al leicl oí, 9 8 1 0 0
lean produciiuli ;1s huiiianii. hililiirig, 100-3
al Ne\v UniieO M o ~ o bllaniilaciui.irig
r
I n c , 82-84, 101
pil>ductiirity in, R3, 85
Autornuiii.~dcsion
cump;il-isun o! lean aiid mass prij
duciiuii, 117-19
comparisori oi rcscür-iii aiid dcvcl<iprneiii in iiiass v c r u lcaii productiiin, 132-34
conscquences ul' lean iiesigii iii malkc~pliicc,i 19-26
l'uturc in, 127-28
iean isiiiovatiuri i i i piacii<:e, 131-32
necd iur i n n o i a t i o n i n , 13.5-17
product dcvelopmciii
nruund ii'ui-Id, 110-11
in lean proiluctir>n lirm, 109-10
Automutivc clcsign (co>?i.)
in mass-production lirm, 104-9
role of invention
in lean production, 129-30
in mass production, 128-20
techniques of lcan design, 1 12-1 7
Bentlcy, 132
Blanking press, 5 1
Bluc-collar workers, 14
BMW. 65, 119, 188, 198
Busch, 164, 166
Boscli, Rohert, 62
Brazil, automobilc industry in, 269-70
British Lcyland, 193
gui.crnmcnt control of, 197
nationalization of, 234
Bromont Hyundai plant, 263
Buick, 105
Buiclt Century, 104
Buick Rcgal, 108
Buick Skylark, 106
Cadillac, 105, 106
limousines, 209
Calvet, Jacqties, 255
CAMMI (GM-Suzuki) plant, 252
Canada-U.S. Auto Pact (1965). 264
Canada-U.S. 1:ree Tradc Agrecment
(1989), 264
Career ladders, in lean enterprisc
managemrnt. 198-200
Center for Tcchnolugy, Policy and Industrial Development, 4
Chandler, Alfrcd, 34
Chanechun.
268
Channcl loyalty, in lean production,
185-86
Chaplin, Charlie, 231
Chevrulet, 105, 106. Sce nlio Gencral
Motors
Chcvrolct Celehrity, 104
Chevrolet Clievelle, 106
Chevrolet Corvair, 106
Chevrolet Vega, 129
China. automobile industry in, 268-69
Chrysler, 197,213
closing of Kenosha (Wis.) plant, 215
crisis a t , 238
forcign opcraliuns of, 208, 242
and parts supply, 156
Pcntastar program, 159
sales of, 227
SI. Louis plant, 244
strategic alliances of, 219
streamlining "f. 244
supply system a t , 139-40
Chryslcr Europe, 121,213
Clirysler LcBaron, 244
CiLroCn.. 121.234
.
Citrocn, André. 44,231
Clark. Kim. 63. 110. 114.115. 155. 164
Clcan sheet, 11 I
Cologne (Gernrany) Ford Plant, 38, 39,
45 741
Communicatiorr, in lean design, 115
1.h.
Componcnts supply, in lean productiun, 146-53
Computer-aided design (CAD), 153
Craft production, 88-91
characteristics of, 24
drawbacks 01.25-26
versus mass production, 29
orgaliization in, 24
overtake oí, by mass production,
227-28
at Panhard ct Levassor (P&L), 21-25
production volume in, 24
techniques of, 12-13
tuols in, 24
in work forcc, 24
Customcr relations. 169-91
and channel loyalty in lean producLion, 185-86
comparison of lean versus mass disrribution, 186-87
and European customer, 175-75
future of, 187-88
information technology and lean
customer relations, 189-91
and lean dcalership, 184-85
and lean production, 178-84
and mass production. 170-75
Cyclicality, prohlem of, and transition
to lean production, 247-51
Daewou, 261,262,268
Dagcnliam (England) Ford plant, 38.
39.45.231.241
Daimlcr, Gottlieh, 21
Dairnlcr-Beriz (Mercedes), 46, 1x8
Decentralizalion, in supply chain,
138-39
Dcming, W. Edwards, 277
Detail-engincering, 156
Dcutsche Bank, 198
~ ~ ~ ~ ~ lcountries,
o p i n gautumobilc industry in, 87,263-70
~ i ~ m o n d - C i(Mitsubishi-Chrysler)
ar
plant, 252
Diclrcnson, Garv, 108
Dic-makiug
lean-riroduction approach to. 116i7
mass-production appruach 10, 116
Dimensional creep, 22
Dodee Davtona, 244
Doni A. 2-61
Dorii. Robert, 105-6, 107, 108
Drucker, Petcr, 11
du Pont, E. l.. 40
du Pont, Pierre, 40
Durant, Wiiiiam, 39-40
East Asia. automobile production in,
267-69
Elcctronic-veliicle tcchnologies, 135
Eilis. Evelvn Henrv, 21-22,23-24.26
Ephlin, D;>nald, 95
Europcan ai~tomobilcindustr),
distribution system in, 175-75
arid globaliration, 213-15
mass production in, 234-35.277
in tlie 19805, 239
production volumc in, 123
product range in, 121-22
transition to lean production in,
253-55
Europcan Economic Community
(EEC), arid Japanese inveslmcnts, 255
European Frec Trade Area, 267
IZxclusive dealing, 170-71, 176
Fairlax (Küns.) Gcneral Motors plant.
96
Fiiurotc, Fay, 28.32
Ferrari, 63
Fiat, 119, 193, 197, 235,257,267,269
Mirafiori plani, 45.40.235
use of tiering by, 165
~ i ~ ~ ninclcan
e , criterprise mana&rncnt, 192-98
~ i s h e Bady
r
Division (GM), 107
Flins (Rcnauil) plant, 46,235
1 1, 138, 192, 21 l . 227-78.
~ " r d Henry,
,
276
and concept »f mass production, 2630
and customer relations, 170
as isolaiionist, 230
and moving asscmbly line, 28
pacilism o[, 35
and par, interchangcability, 27-28
h r d , Henry, 11, 139, l92-93,211
Ford Anglia, 21 1
Ford Capri, 27 1
Ford CDW 27,212
Ford Escort, 212,213
Ford Fcstiva, 212,262
Ford "Flying Flivvcr," 39
Ford Model A, 26. 39.210
Ford Model T. 26, 27.30, 35, 37.38.
126,210,227,231
Ford Modcl Y, 39,209-10
Ford Motor Company, 12
Atlanra assembly plaiil, 96
Coloene (Germany) piant, 38.39.43.
y4 1
Dagenham (England) plant, 38,39.
45,231,241
Europcan operations u[, 220
financing ol, 192-93
Hern~usillo(Mcrico) planl, 87,26566
Flighland Park plant, 29,30-31, -7637.38.42.62,82,229,23-34
link w k h Mazda, 212-13,237,271
1980s crisis at, 237-38, 244
parts dciign a l , 58-19
performance oi, 244
popularity of carly cai-s. 37-38
Ql program, 159
Rouge complex (Detroit), 33,38-39,
sales of, 227
~ u p p l ychain in carly, 139
Trafford Park (England) planl. 228-
318
INDEX
Ford Motor Company ícori~.)
and use ofoutsourcing, 157
wage leve1 a l early, 42
Ford Sable, 108,212
Ford Sierra, 212
Ford Taurus, 96,107,108.212
Ford Tempa. 212
Ford Topaz, 212
Ford TriMotor, 39
Ford V8,39
Ford Valve Grinding Tool, 30
Framingham (Mass.) General Motors
assembly plant, 77-78,85-86,
244
Fuji Hcavy Industries, 196
Fujimoto, Takahiro, 110-1 1, 114, 115,
155,164
General Electric Plastics, 162
General Motors, 12.37. 155, 257
A-cars, 104.108-9
Corvair project, 129
decentralization a t , 41
Fairfax (Kansasj assemhly plant, 96
hve-modcl product range a t , 41
foreign operations of, 242
founding of, 39
Framingham (Mass.) assembly
piant, 77-78,85-86.244
G-cars, 104
GM-10 project. 96, 104-9, 118, 129
joint ventures of, 219,238,267
in 1980s. 238-39
outside financing for, 41
parts design a t , 58-59
problem-solving by Sloan in early,
GM-10 project, 96, 104-9, 118, 129
Goudevert, Daniel, 274
Graves. Andrew. 5
Great Britain, mass production in,
228-31
Greenhouse effect, 137
Handclsbank, 198
Harrison Radiator, 138
Heiju~ika,in lean suppiy, 151
Hermosillo (Mexico) plant, of Ford Motor Company. 87.265-66
Highland Park Ford plant. 29.30-31,
36-37,38,42,62,82,229,23134
Honda
distribution channels of, 180
cngineering design at, 129-30
European base of. 216-17
~ a r y s v i i l e(Ohioj plant, 116-17,
198-99,201,202,241
North American opcrations of, 201,
216-17,241
and use of multiregional production,
7n5-7.
and United Auto Workers, 252-53
Honda Accord, 108, 109-10,202,218
HondaIAcura Lcgend, 216
Honda ConcertoIRover. 200
Honda NS-X, 25.65
Hubai, 268
Hyatt Roller Bearing Company, 40
Hyundai, 261-63,268
Bromont (Quebec) plant, 263
Hyundai Excel, 261-62
19-47
.
. .-
and retraining o1 workers, 259-60
sales of, 227
Spcar program of, 159
strcamlining of, 244
supply chain in early, 138-39
supply system at, 139-40
Vega projcct, 129
X-car project, 129
Geographic sprcad, in lean entcrprise
management. 200-3
GKN. 164
Global enterprise
advantages of, 204-9
managing, 209-13
Iacocca, Lee, 219
Indirect workers, in mass-production
plantc, 78
Industrial engineers, 31, 32
Information tcchnology, and lean customer relations. 189-91
Interchangeablity, 27,28
International Motor Vehicle P r ou~ r a m
(IMVP), 4-7
attendees at policy forums, 7, Appcndix C
organirations contributing to. 6, Appendix A
INDEX
researcli smfl, 6, Appendix B
World Assemblv Plant Survey, 75%
Invcntion
in Lean production, 129-30
in mass production, 128-29
Inventory
in mass production, 78
in lean production, 62,80,95, 160.
61
Invisible hand, 34
lsuzu. 196,208,219
Jaguar, S . 1 19,166,204
Japanese autornohile industry. Ser also
Hundü: Mazda: Mitsubishi; Nissan; Toyota
autoniobilc dealerships in, 184-85
and brand loyalty, 18-5-86
enginc cvolution in, 131-32
llcxible compensation in, 250
global stratcgies of, 215-18
hiring policies in, 251
implications hehind U.S. investmcnts of, 252-53
investmcnt of, in U.S. auto industry,
240-42
inward focus 01,272-75
lifetime cmployment in, 54
production volume in, 123
pruduct range i r i , 119, 121
quality circles in. 56
rcwork in, 56
scniorits systcni in, 54
Japanese Ministry of Internationai
Trade and Industry (MITI), 5051
Job-control unionism, 42
Just-in-time inventory, 62,95, 160-61
Kaizeiz, in lean supply. 149-50
Kamiyam, Shotaro. 66
Kanban system, 62.67
Kansai Kyohokai, 153
Kanto lcyohokai, 153
Kenoshu (Wis.) plant, ciosing of, 215
Keiretsii systern in Japan. 194-95.274
Kia, 261, 262, 268
Knowlcdge workers, 32
Koito, 61, 195
Korean auto industry. 261. See also
Hyundai
Korean Ministry of Industry, 261, 262
Krafcik, John, 5, 13,751'. 77,82
Lamming, Richard, 5, 156
Leadership, in lean design, 112-13
Lean customer reiations
loture of, 187-88
and information technology, 189-91
Lean dealership, 184-85
Lean design
conseauences of, in marketplace,
1i9-26
techniques, 112
communications in, 115-16
leadcrship, 112-1 3
simultaneous development in
116-17
tcamwork in, 113-15
Lean enterprisc, 73
Lcan cntcrprise management
and advantages of global cnterprise,
204-1 3
career ladders in, 198-200
failure of European industry in,
213-15
finance in, 192-98
geographic spread in. 200-3
Japancse in, 215-18
and multiregional enterprise. 21822
Lean organization, at thc piant le\'el,
98-1 O0
.ean production
hirthplace o[. 49-51
and changing consumer demand
64-65
channel loyalty in, 185-86
company a s comrnunity, 53-55
components supply in, 146-48
comparison ofmass production
with. 80-82
in contrast tu n>iissproduction, 13-~
14, 186-87
and customer relations, 178-84
and dcaline
- with the custorner, 6668
diffusing, 82-84
encountcr between mass production
and, 235-36
1.ean productiun iconr.)
final assembly plant in, 55-58
future of. 68-69
growth o l 12
as hurnanl~rluliilline.
-. 100-3
initial misperceptions of, 236
innovation in practice, 131-32
imrention in, 129-30
inventory in, 80
making iransitions 10,257-75
obstacles lo, 257-75
pr«duct developmeni and engineering, 63-64, 109-10
quality control in, 79
research and úcvelopmcnt in, 13234
supply chain in, 58-62
techniques oí, 13.51-53
at Toyota Motor Company, l l . 4869,79-80
transition to, in Europea" auto industry, 253-55
work forcc in, 53-55,80
Lean supply, in practice, 148-53
Leland, Henry, 36
Levassor, Emile, 21.22
Lex Group, 175, 176
I.iktimc employment, 54
Lucas, 166
MacDufTic, John Paul, 75ri
Manufacturiny engineers, 32
Marelli, Magneti, 166
Markctplace, consequenccs of lcan desien in ihc. 119-26
Marvsviile (Ohio) Honda plant, 11617, 198-99.201.202.241
Mass production
in America in 1955,43
British systcm oí, 228-31.233-34
comparison ul, with lean pi-oduclion, 80-82
i i i contras1 tu lean pr-oduciioii, 1314, 186-87
versus ci-aft production, 29
and customer relations, 170-75
diffusion oí, 44-47
encuunter betwecn Lean ~ r o d u c t i o n
and, 235-36
in Eumnc, 234-35.277
Fí~rd'sconcept 01.26-30
at GM Framingham plant, 77-78,
85-86
invention in, 128--29
loeical
limits of. 38-39
.
and opposition to lean production,
257-60
orgunizations in, 33-35
overtake of craft production by,
227-28
parts design in, 140-42, 156-62
parts snpply in, 142-46
and product develupment, 104-9
prodnct in, 37-38
research and develupment in, 13234
risc and fall o[, 21-47
techniuues of. 13
1001s iri, 35-37
work iorce in, 30-33.55.78
Mazda, 68, 186,208. 194,252
distribution cliannels o[, 180
íalniiy management o[, 196-97
lean production at. 237
link with Ford Motor, 212-13.237
27 1
North American operations 01,201
Mazda Miata, 213
McKenna tariíí, 228
Mercedes-Benr, 2 l , 6 5 , 119. 198
Mercury Sable, 96
Mercury Tracer, 265
Mcxican Auto Decree, 270
Mcxico, auinmohilc industry in, 87,
240-41.264-67
MG, 46,231
Michelin familv. 197
Miraliori ( ~ i a t f p l a n t45,46,235
.
Mitsuhishi Motor Company. 194
distribution channels of. 180
North American operations os, 201
strategic alliances ol, 219
Miyoshi,Tateomi. 109
Morris, William, 44, 231-33, 234
Moiorola, 162
h.Iultircgional Motors (MRM), 21 8-22
My Eurs tvirh Gerrerai Morors (Sloan),
128
Nash, 171
Neocraftsmanihip, 101-2
New Unitcd Motor Manulacturing Inc.
(NIJMMI) plrint (I'remorlt. Calif.), 5 7 7 , 82-84, 219, 252
;inri 5unolier
oerfo~-mance,163
,.
success of, 238-39
Nippondenso, 61.62. 195
Nishiguchi, Toshihiro, 5
Nissan, 51
distribulion channels oí, 180
Mexican olant o[. 240-4 1
Nortli American operations oí, 201.
202-3.241.265
regional supplier associations "f.
153
and Unitcd Auto Workers, 252-53
Nissan Infiniti UJ5, 146,209
Nissan Sentra, 203
Nisshin Kogvo, 146
Nobcoka, Kentaro, 6. 112
NS-X sports car (Honda), 25.65
.
Oldsmobile, 105, 106
Oldsmobile Ciera, 104
Oldsmobile Cutlass Supreme. 108
Oldsmubile 1;-85,106
Ohnri. Taiichi. 11.49,51.52,56, 62, 63.
67,235,277
Opel, 234,242
Opcl Kadett, 262,267,270
Organizalion
in craft productlon, 24
in mass production, 33-35
Outsourcing. 157
Oxford Motor Company, 23 1
Panhard et Lcvassor (P&L). 22, 126
craít-production system a t , 21-25
Pengeot, 121, 165
Peugeut Family, 193,197
Pickcns, T. Boone, 195
Pioncer Electric, 5
Pontiac,98,105
Pontiac Grand Prix, 96, 108
Pontiac LeMans, 262
Pontiac 6000,104
Puntiac Tempest, 106
Porsche. 58.198
PorschefPeicli lamil?, 198
Prehardencd melals, 27
Produci development
and lcan production, 109-10
and m a s ~
pi-oduction, 37--38, 104-9
in U.S. au~omoti\,eindustry, 245
woridwide, 110-1 1
Product engincers, 32
Pruductiun srnoo~hiiig,i i i lean suj>plv,
151
Production vuiumc, in craSi pruduction, 24
PSA, 119, 121, 197
Quality, 55
Qualitv circles, 56
Quality control, i n lean producti<iri,79
Quandt hrnily, 198
Rcnault, 119, 121, 171, 193,213,235,
257
North American uperations of, 215
ourchasc/sale oí American Motors
by, 214
use oí tiering by, 165
Renault, Louis, 44,231,234
Rescarch and dcvclopmeni in lean
production versus mass productiori, 132-34
Kooi, Dariiel, 4
Rouge Ford iomplcx (Dctroii). 33,3839.48.49
Rovrr, 119, 193,216,134
Saab, 58, i i 9 , 121,204,198
supply chain fui-, 139
Saginaw Steering, 138
Saint Louis Chrysler plani, 244
Schmidt, Püul, 108
Seat, 121, 193
Seiki. Aisliin, 61
Seniority, 54
Sherilí, Antony, 6,i 12
Shohokai, 153
Shup-floor worker, 33
S / l ~ s asystem, 112-15. 146-48
Sicmens, 162
Simultaneous dri,cloprncni, in Icün de.
sign, 116-17
Sinsle-sourcinri, 157-59
Smith, Adam, 34
Smith, Jack, 238
Society of Industrial Engineers, 153
Solex, 166
Statisiical process control (SPC), 153,
159-60
Steyr. 213
Studehaker, 171
Subaru, 186
Sumitomo, 68, 194. 197
Supply chain, 138-67
bidding in, 139, 140
components supply in lean production, 146-48
and cost adjustmcnts, 141
and debugging, 144
and dccentralizution, 138-39
heijunku in, 151
hurdles in lean supply, 167-68
kaizen in, 149-50
lean supply in practicc, 148-53
making running changes in, 144
parts design in mature mass production, 140-42
parts supply in mature mass produciion, 142-46
and prohlcm of fluctuating volume,
145
and production smoothing. 15 1
and quality control, 140-41. 144
reforming mass-production supply
systems, 156-62
supplier
associations in lean sunnlv.
..
.
153-56
in U.S. automotive industrv, 244-45
and value engineering, 148-49
Western Europe as halfway station,
163-67
and zero defects, 151-52
Suzuki, 219
Systeme Panbard, 21
..
Tukarakai, 153
Teamwork, in lean design, 113-15
Tokai Kvohokai. 153
Tools
in craft ~ r o d u c t i o n24
.
in mass production, 35-37
Tushihiro Nishiguchi, 162
Total quality control (TQC), 153
Toyoda. Kiichiro, 48, 53, 54
Toyota, Eiji, 11.48-49,62,63,66,231,
235,277
Toyota Auto, 180
'h"ota Camry, 180
Toyota Celica, 180
Tuyota Corolla, 180-82
Toyota Gosei, 61, 19.5
Toyom-GM joint-venture plant
(NUMMI). See New United Motor Manufacturing Inc.
(NUMMI) plant
Toyota Lcxus LS400,209
Toyota Motor Company, 51.66-67, 194
aggressivc selling by, 67-68
components supply for, 60-62
distribution channcls of. 180-81
efficicncy ol. 155
Europcan operations of, 241
"five why's" philosopby a t , 57, 152
Georgetown (Ky.) plant, 119
joint venturc with General Motors,
238
lean production a t Takaoka plant,
79-80
1946 strikc at, 53-54
North American operations ol, 201
pionecring of lean production a t , 11
and quality control, 59-60
regional supplier associations "f.
153
rise of lean production at, 48-69
shusa system at, 112-15
spin-off ofsupplicr companies by,
195
.,-
and United Auto Workcrs. 252-53
Toyota Production System, work pacc
in. 259
Toyota Publica, 180
Toyota Takaoka plant
accuracy a t , 81
productivity a t , 81
Toyota Supra, 180
Toyota Toyopet, 180
Toyota Townall van, 180
Toyota Vista, 180
Trahani Volkswagen plant, 267
Trafford Park (England) Ford plani,
228-29
UAW. See United Auto bvorkcrs
Udevalla systcm, 101-2
Unibody car, 46
United Auto Workers (UAW),42,25253
and lcan production, 83. 100-1
United Kingdom, brand loyalty in, 185
U.S. automohile industry. See also
Chrysier; Ford Motor Cumpany;
General Motors
companies in, 227
diffusionof leari production a l , 24445
employment in, 247
Jawancsc investmeni in, 240-42,
252-53
and problem ofcyclicality, 247-51
pi-oduci development i n , 245
production volume in, 123
product rangc in, 119, 12 1
supply system at. 244-45
transition to lean production, 246
U.S. car dcalerships, 171-75
hazaar tradition in, 173-74
changes in, 171
inventory in, 171
Utilitv man, 55
Visible liand, 34,38
Volkswagen, 119, 171, 193,735,237,
274
Norili American operations of, 21314,215
Traban1 piant, 267
Wolfsburg plant, 45.46.235
Volkswagen Bceile, 46
Volkswagen Fox, 269.270
Volkswagcn Polo, 267
Volvo, 119, 121, 198
distribution system of, 173
Udevalla plant, 101-2
Volvo Concessionaires, 175, 178
Wallenberg lamily, 198
N'ankel rOtaTV eneine, 196,237
~ a r t h u r g2, 6 j Wolfburg (VW) plant, 46,235
~ o r force
k
in craft production, 24
in lean production, 53-55.80
iri mass production, 30-33,s;. 78
Yield, 55
Valeo group, 166
Value analysis (VA), 153, 160
Valuc engineering (VE), 153
in lcan supply, 148-49
Vertical integration. 33,34,58
Zaibutsu system in Japan, 193-94
Zcro defects, as goal in lean supply,
151-52